JP2019149140A - Information search device and information search system - Google Patents

Abstract

To provide an information search device which can detect a search target more accurately.SOLUTION: An information search device 2 includes: a positional information acquisition unit 22 for acquiring positional information; a category determination unit 23 for determining the category of a search phrase; a feature word generation unit 24 for generating a feature word related to a word indicating a search target from the search phrase; and a search unit 25 for searching for the search target from a storage database. The search target has point information. The search unit 25 searches for the search target corresponding to the feature word in a predetermined range from the position and in the determined category on the basis of the positional information, the category, and the feature word.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information search device and an information search system, and more particularly, to an information search device and an information search system that search for a search target representing a specific point based on a search phrase.

  2. Description of the Related Art Conventionally, in map search, there is known an information search apparatus capable of performing map search not only on data in which coordinate information is explicitly registered but also on information included in the contents of atypical data (Patent Literature). 1).

  In the information search device of Patent Document 1, data including a plurality of words having attributes is input, a plurality of words included in the input data are extracted, and attribute-related words obtained by extracting a plurality of words for each attribute are converted. Searching is performed using data to which attribute information is added.

  For example, in the case of the word “Hokkaido” relating to the position attribute, the information search device of Patent Document 1 is based on the search keyword input by the user because the coordinate information of “Hokkaido” is added to the atypical data. The predetermined search result can be output.

JP 2014-191714 A

  However, in the configuration of the information search apparatus of Patent Document 1, a search result cannot be output unless the word has an attribute, and a search target with higher accuracy cannot be detected from the search phrase.

  An object of the present invention is to detect a search object with higher accuracy.

  The information search device according to the first aspect of the present disclosure includes a position information acquisition unit, a feature word generation unit, and a search unit. The position information acquisition unit acquires position information. The position information indicates the position on the map that is the starting point of the search. The feature word generation unit generates a feature word from a search phrase. The feature word is related to a word representing a search target. The search unit searches the search target from a storage database. The accumulation database accumulates the feature words and the search target. The search target corresponds to the feature word. The search target has point information. The point information indicates a specific point. The search unit searches the search target corresponding to the feature word within a predetermined range from the position based on the position information and the feature word.

  The information search device of the first aspect can detect a search object with higher accuracy.

  In the information search device according to the second aspect of the present disclosure, the position information acquisition unit acquires the position information. The position information is estimated from words included in the search phrase by machine learning.

  The information retrieval apparatus of the second aspect can estimate position information by machine learning. The information search device according to the second aspect can estimate the position information indirectly from the words included in the search phrase even if the words indicating the position information directly are not included in the search phrase.

  The information search device according to the third aspect of the present disclosure further includes a category determination unit and a search storage device. The category discriminating unit discriminates a category of the search phrase. The search storage device stores a word corresponding to the category. In the accumulation database, the feature words are classified for each category. The category determining unit includes a category determining unit and a category estimating unit. The category determination unit determines the category based on whether or not a word included in the search phrase matches a word stored in advance in the search storage device. The said category estimation part estimates the said category from the word contained in the said search phrase which was not determined by the said category determination part by machine learning. The search unit searches the search object corresponding to the feature word in the predetermined range and the determined category based on the position information, the category, and the feature word.

  The information retrieval apparatus of the third aspect can estimate the category by machine learning. The information search device according to the third aspect can estimate the category indirectly from the word included in the search phrase even if the word indicating the category directly is not included in the search phrase.

  In the information search device according to the fourth aspect of the present disclosure, the feature word generation unit has a higher similarity to the word representing the search target from words including the noun, the adjective, the verb, and the adverb in the search phrase. Generate feature words.

  The information search device according to the fourth aspect can detect a search target with higher accuracy by generating a feature word having a higher similarity to the word representing the search target.

  An information search system according to a fifth aspect of the present disclosure includes an information processing terminal, an information search device, and an information storage device. The information search apparatus is configured to be able to communicate with the information processing terminal. The information storage device is configured to be accessible by the information search device. The information processing terminal includes an input device, a terminal communication device, and a display device. The input device is configured to input a search phrase. The terminal communication device is configured to be able to transmit the search phrase to the information search device. The terminal communication device is configured to receive a search object searched based on the search phrase from the information search device. The display device is configured to display the search target.

  The information search device includes a search communication device, a position information acquisition unit, a feature word generation unit, and a search unit. The search communication device is configured to receive the search phrase from the information processing terminal. The search communication device is configured to transmit the search target. The position information acquisition unit is configured to acquire position information. The position information indicates the position on the map that is the starting point of the search. The feature word generation unit generates a feature word from the search phrase. The feature word is related to a word representing the search target. The search unit is configured to be able to search for the search target from a storage database.

  The search target has point information. The point information indicates a specific point. The search unit is configured to search the search object corresponding to the feature word within a predetermined range from the position based on the position information and the feature word. The information storage device includes a storage device. The accumulation storage device stores the accumulation database. The accumulation database accumulates the feature words and the search target. The search target corresponds to the feature word.

  The information search system of the fifth aspect can detect a search object with higher accuracy.

  In the information search system according to the sixth aspect of the present disclosure, the information processing terminal further includes a terminal storage device. Map information is stored in the terminal storage device. The display device displays the specific point indicating the search target on a map of the map information.

  The information search system according to the sixth aspect can display a search target point with higher accuracy on a map.

  In the information search system according to the seventh aspect of the present disclosure, the feature word generation unit calculates statistical data of the search phrase. The information storage device associates the search phrase received by the search communication device with the statistical data within a preset period from the information processing terminal, and stores it in the storage database.

  The information search system of the seventh aspect can detect a search object with higher accuracy based on a search phrase within a preset period.

  In the information search system according to the eighth aspect of the present disclosure, the feature word generation unit generates the feature word having a higher degree of similarity with the word representing the search target from the words included in the search phrase. The feature word generation unit preferentially outputs the feature words having a higher frequency based on the statistical data.

  The information search system of the eighth aspect can detect a search object with higher accuracy.

  According to the present invention, it is possible to detect a search object with higher accuracy.

It is a schematic block diagram of the information search system which concerns on this embodiment. It is a flowchart figure which shows the processing operation of the information search system which concerns on this embodiment. It is a flowchart figure which shows an example of the process which searches a spot based on a cluster. It is a figure which shows the specific example which determined the cluster from the search phrase. It is a figure which shows an example of the data table stored in the category database. It is a figure which shows an example of the data table stored in the spot database. It is a block diagram which shows an example of a principal part structure of an information storage device. It is a flowchart figure which shows the example of the process which produces | generates a spot database. It is a flowchart figure which shows the example of the process which produces | generates a category database.

  Hereinafter, the information search system 10 of this embodiment will be described. As shown in FIG. 1, the information search system 10 includes an information processing terminal 1, an information search device 2, and an information storage device 3. The information search device 2 is configured to be able to communicate with the information processing terminal 1. The information storage device 3 is configured so that the information search device 2 can access it.

  The information processing terminal 1 includes an input device 11, a terminal communication device 12, and a display device 13. The input device 11 is configured to input a search phrase. The terminal communication device 12 is configured to be able to transmit a search phrase to the information search device 2. The terminal communication device 12 is configured to be able to receive the search target searched based on the search phrase from the information search device 2. The display device 13 is configured to display a search target.

  The information search device 2 includes a search communication device 21, a position information acquisition unit 22, a feature word generation unit 24, and a search unit 25. The search communication device 21 is configured to receive a search phrase from the information processing terminal 1. The search communication device 21 is configured to transmit a search target. The position information acquisition unit 22 is configured to acquire position information. The position information indicates the position on the map that is the starting point of the search. The feature word generation unit 24 generates a feature word from the search phrase. The feature word is related to a word representing the search target. The search unit 25 is configured to be able to search for a search target from the accumulated database.

  The search target has point information. The point information indicates a specific point. The search unit 25 is configured to be able to search for a search target corresponding to a feature word within a predetermined range from the position based on the position information and the feature word.

  The information storage device 3 includes a storage device 31. The accumulation storage device 31 stores an accumulation database. The accumulation database accumulates feature words and search targets. The search target corresponds to the feature word.

  The information search system 10 of this embodiment can detect a search target with higher accuracy.

  Hereinafter, each structure in the information search system 10 of this embodiment is demonstrated concretely.

  The information search system 10 according to the present embodiment is configured such that the information processing terminal 1 used by the user and the information search device 2 can communicate with each other. The information processing terminal 1 and the information search device 2 are connected by wireless communication via the public communication network 50, for example. In FIG. 1, the flow of information that is communicated via the public communication network 50 is illustrated by double arrows.

  As illustrated in FIG. 1, the information processing terminal 1 includes a terminal control device 14 and a terminal storage device 15 in addition to the input device 11, the terminal communication device 12, and the display device 13. The information processing terminal 1 is a terminal device used by a user. Examples of the terminal device include a mobile phone or a smartphone.

  The input device 11 is configured to input a search phrase. The search phrase is a word for searching for a search target. The search phrase is also a word arbitrarily input by the user. The search phrase may be a word, a combination of words, or a composed sentence including a subject and a predicate. The input device 11 is configured to instruct transmission of an input search phrase from the terminal communication device 12. Examples of the input device 11 include a touch panel integrated with the display device 13, a voice input device capable of inputting voice, or a physical key. The input device 11 is not limited to the configuration of the touch panel configured integrally with the display device 13, and may be configured only by the input unit.

  The terminal communication device 12 is configured to be able to communicate with the information search device 2. The terminal communication device 12 transmits search phrase information. The terminal communication device 12 receives the search target information. Since the search target has information indicating a specific point, it is also point information. The point information is POI (Point Of Interest) information. Specifically, the terminal communication device 12 is configured to be able to communicate with the search communication device 21 of the information search device 2 using the public communication network 50. The terminal communication device 12 may be configured not only for public communication using the public communication network 50 but also for wired communication, for example. Further, the terminal communication device 12 may be configured to be able to perform wireless communication corresponding to a communication standard of infrared communication, Wi-Fi (registered trademark), Bluetooth (registered trademark), or BLE (Bluetooth Low Energy).

  The display device 13 is configured to display various types of information. The display device 13 displays a search phrase, for example. The display device 13 may display a map. The display device 13 may display the spot indicated by the search target spot information searched by the information search apparatus 2 so as to overlap the map. For example, the display device 13 can visualize a plurality of pieces of point information and display the information on a map indicated by the map information stored in the terminal storage device 15. The display device 13 is not limited to the configuration of the touch panel configured integrally with the input device 11, and may be configured only by the display unit.

  The terminal control device 14 controls various information processing operations. The terminal control device 14 is electrically connected to the input device 11, the terminal communication device 12, the display device 13, and the terminal storage device 15. The terminal control device 14 is configured to control the input device 11, the terminal communication device 12, and the display device 13 based on a program stored in the terminal storage device 15. The terminal control device 14 is configured by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit).

  The terminal storage device 15 has a volatile storage device whose information disappears when the power is turned off and a non-volatile storage device whose information does not disappear even when the power is turned off. The volatile storage device temporarily stores information processed by the terminal control device 14. The volatile storage device stores, for example, map information indicating a map and point information added to the map information. Examples of the volatile storage device include a RAM (Random Access Memory). The nonvolatile storage device is a storage capable of storing various information processing programs and various information. Examples of the nonvolatile storage device include a ROM (Read Only Memory). For example, a flash memory or an HDD (Hard Disk Drive) is used as the ROM.

  As shown in FIG. 1, the information search device 2 includes a search communication device 21 and a search control device 27. The information search device 2 includes a search storage device 28 in addition to the search communication device 21 and the search control device 27. The information search device 2 is configured by a server, for example. The search control device 27 includes a position information acquisition unit 22, a category determination unit 23, a feature word generation unit 24, a search unit 25, and a timer unit 26.

  The search communication device 21 is configured to be able to communicate with the information processing terminal 1. The search communication device 21 is configured to receive information on the search phrase from the information processing terminal 1. The search communication device 21 is configured to be able to transmit the search target information searched by the search control device 27 to the information processing terminal 1. That is, the search communication device 21 receives information on the search phrase. The search communication device 21 transmits information to be searched. Specifically, the search communication device 21 is configured to be able to communicate with the terminal communication device 12 of the information processing terminal 1 using the public communication network 50.

  The search communication device 21 is configured to be able to communicate with the information storage device 3. The search communication device 21 transmits information for searching for a search target. The search communication device 21 receives the searched search target information from the information storage device 3. Specifically, the search communication device 21 is configured to be able to communicate with the storage communication device 33 of the information storage device 3 using the public communication network 50. The search communication device 21 may be configured not only for public communication using the public communication network 50 but also for wired communication, for example. Further, the terminal communication device 12 may be configured to be able to perform wireless communication corresponding to a communication standard of infrared communication, Wi-Fi (registered trademark), Bluetooth (registered trademark), or BLE.

  The search control device 27 is electrically connected to the search communication device 21 and the search storage device 28. The search control device 27 is configured such that the position information acquisition unit 22, the category determination unit 23, the feature word generation unit 24, the search unit 25, and the time measuring unit 26 function. The search control device 27 is driven to operate, for example, a position information acquisition program, a category determination program, a feature word generation program, a search program, and a time measurement program stored in the search storage device 28. The search control device 27 is composed of, for example, a CPU or MPU.

  The position information acquisition program functions as the position information acquisition unit 22 that acquires position information by being executed on the computer. The category discrimination program functions as a category discrimination unit 23 that discriminates the category of the search phrase by being executed on the computer. The feature word generation program functions as a feature word generation unit 24 that generates a feature word from a search phrase by being executed on a computer. The search program functions as a search unit 25 that searches a search target from a storage database stored in the information storage device 3 by being executed on a computer. The timekeeping program functions as the timekeeping unit 26 capable of timekeeping the year, date, and time by being executed on the computer.

  The position information acquisition unit 22 acquires position information. The position information is, for example, longitude information and latitude information. The position information may include altitude information in addition to longitude information and latitude information. The position information acquisition unit 22 can acquire position information transmitted from the information processing terminal 1, for example. The position information transmitted from the information processing terminal 1 includes, for example, position information input by the input device 11 of the information processing terminal 1. Further, the position information acquisition unit 22 can acquire position information from a search phrase transmitted from the information processing terminal 1.

  The search phrase is composed of text data. The search phrase includes a phrase. The phrase may be a single word or a group of a plurality of words. The phrase may be a written sentence. The search phrase is converted into a format in which text data can be handled mechanically. The search phrase is decomposed into a plurality of words by morphological analysis. Here, the morphological analysis can classify the search phrase into the smallest units that have meaning in natural language. In the morphological analysis, the part of speech information of the classified morphemes can also be acquired. In morphological analysis, for example, it is also possible to acquire information about whether a word included in a search phrase is a noun, an adjective, a verb, or an adverb.

  For the morphological analysis, for example, a morphological analysis engine such as MeCab can be used. MeCab can be used with a named entity dictionary. An example of the specific expression dictionary is a system dictionary such as Neologd. The position information acquisition unit 22 performs morphological analysis on the search phrase in order to obtain a word included in the search phrase. The position information acquisition unit 22 is not limited to a configuration that morphologically analyzes a search phrase in order to obtain a word included in the search phrase. The position information acquisition unit 22 may have a configuration in which, for example, the category discrimination unit 23, the feature word generation unit 24, the search unit 25, or a morpheme analysis unit (not shown) obtains a word obtained by morphological analysis of the search phrase. The position information acquisition unit 22 may perform a distributed expression in which a word subjected to morphological analysis is converted into a numerical expression. The position information acquisition unit 22 may collect utilization forms of words subjected to morphological analysis. For example, FastText may be used in order to summarize the utilization forms of words subjected to morphological analysis.

  When the word included in the search phrase is a place name, the position information acquisition unit 22 can obtain position information from a place name database in which the place name and the position information are associated with each other. The position information acquisition unit 22 compares, for example, a word obtained by decomposing the search phrase by morphological analysis with a place name database stored in the search storage device 28. If the word obtained by decomposing the search phrase by morphological analysis matches the place name in the place name database, the position information acquisition unit 22 may acquire position information corresponding to the matched place name.

  The place name database may be configured by associating names of prefectures, cities, wards, towns, villages, or facilities with location information, for example. Here, the place name includes not only the name representing the city, but also the name of the facility. In other words, the place name is a name indicating geographical information. In the place name database, for example, the name of the prefecture and the position information of the prefectural office, the name of the city and the position information of the city office, the name of the ward and the position information of the ward office, the name of the town and the position information of the town office, the name of the village and the village office The location information, the name of the facility, and the location information of the facility correspond to each other. Hereinafter, when the word included in the search phrase is a place name, obtaining the position information from the place name database in which the place name is directly associated with the position information is also referred to as obtaining position information by non-machine learning. When the position information acquisition unit 22 acquires position information by non-machine learning, the position information acquisition unit 22 performs a hard search as to whether or not it matches a word included in the database.

  When the word included in the search phrase does not include a place name, the position information acquisition unit 22 can acquire the position information from a database in which the search word and the position information are associated by machine learning. The position information acquisition unit 22 replaces the search phrase with a word that has been decomposed by morphological analysis and a search word having a high degree of similarity, for example, by machine learning. For example, the position information acquisition unit 22 can perform a morphological analysis on the search phrase and then select a desired search word based on the similarity between the word stored in advance and the search word. The position information acquisition unit 22 compares the search word with a place name database in which the position information is associated. If the search term matches the place name in the place name database, the position information acquisition unit 22 may acquire position information corresponding to the matched place name. In other words, the position information acquisition unit 22 uses a geocoding technique for converting a place name indicating geographical information into coordinate values of longitude and latitude.

  The position information acquisition unit 22 uses at least one of position information from the information processing terminal 1, a place name word included in the search phrase from the information processing terminal 1, and a search word based on the search phrase from the information processing terminal 1. Thus, position information indicating the position on the map can be acquired. The position information acquisition unit 22 uses the information from the information processing terminal 1 to acquire position information indicating the position on the map. In the information search device 2 according to the present embodiment, even when the position information cannot be acquired directly from the information processing terminal 1, the position information can be acquired indirectly. The information search device 2 can acquire position information with a lower processing load compared to a configuration that acquires position information only by machine learning.

  That is, in the information search device 2 according to the present embodiment, the position information acquisition unit 22 acquires position information. The position information is estimated from the words included in the search phrase by machine learning.

  The information search device 2 according to the present embodiment can estimate position information indirectly from words included in a search phrase by machine learning even if a word directly indicating position information is not included in the search phrase. it can.

  The category determining unit 23 is configured to be able to determine the category of the search phrase. The category determination unit 23 determines the category of the search phrase transmitted from the information processing terminal 1. Categories include, for example, hot springs, parks, museums, shrines, live performances, shopping, amusements, outdoors, events, gourmet, ramen, and mountains. Categories are divided into appropriate types. The category type can be set to 300 types, for example.

  The category discrimination unit 23 performs morphological analysis on the search phrase. The category discriminating unit 23 is not limited to a configuration that performs morphological analysis of the search phrase in order to obtain a word included in the search phrase. For example, the category determination unit 23 may be configured to acquire a word obtained by morphological analysis of a search phrase by a position information acquisition unit 22, a feature word generation unit 24, a search unit 25, or a morpheme analysis unit (not shown). The category discriminating unit 23 may perform a distributed expression in which the words subjected to morphological analysis are expressed in numerical values. The category discriminating unit 23 may collect utilization forms of words subjected to morphological analysis.

  When the word included in the search phrase is a category name representing a category, the category determination unit 23 can directly determine the category from the category database in which the category name and the word are associated with each other. The category discriminating unit 23 compares, for example, a word obtained by disassembling the search phrase by morphological analysis with a category database stored in the search storage device 28. If the word obtained by decomposing the search phrase by morphological analysis matches the category name in the category database, the category determination unit 23 may store the matched category name in the search storage device 28 in association with the search phrase. Hereinafter, when the word included in the search phrase is a category name, determining the category from the category database is also referred to as category determination. That is, it can be said that the category determination unit 23 includes a category determination unit 231 that performs category determination by non-machine learning from a category database when a word included in a search phrase is a category name based on category determination.

  When the word included in the search phrase does not include the category name, the category determination unit 23 can estimate the category from the category database in which the search word is associated with the category name by machine learning. The category discriminating unit 23 replaces a word obtained by decomposing a search phrase by morphological analysis with a search word having a high degree of similarity with the word, for example, by machine learning. In this case, the category discriminating unit 23 may replace one word with a search word having a high similarity to the word among the words obtained by decomposing the search phrase by morphological analysis, or replace two or more words with those words. It may be replaced with a search term having a high similarity. For example, the category determination unit 23 may extract a co-occurrence word that co-occurs from two or more words, and replace the co-occurrence word with a search word having a high similarity to the words that are clustered in advance for each category. . Here, co-occurrence means that a predetermined word included in a search phrase appears simultaneously with another word in the search phrase. A co-occurrence word is another word that appears simultaneously with a predetermined word. Since categories often use different words depending on the classification, it is preferable that the category determination unit 23 performs category determination by machine learning using teacher data whose output corresponding to the input is known in advance.

  For example, the category determination unit 23 can select a desired search word based on the similarity between the word stored in advance and the co-occurrence word after performing morphological analysis on the search phrase. The desired search term is set to match the category name. The category discriminating unit 23 compares the search term that matches the category name with the category database. If the search term matches the category name in the category database, the category discriminating unit 23 may estimate that the matched category name is the category of the search phrase. That is, the category discriminating unit 23 includes a category estimation unit 232 that performs category estimation from the category database by machine learning when a category name is a search word similar to a word included in the search phrase based on category estimation. I can say that.

  The category determination unit 23 determines a category by using a word indicating a category included in the search phrase from the information processing terminal 1 or a search word based on a word included in the search phrase from the information processing terminal 1. Can do.

  In the information search device 2 according to the present embodiment, even when the category cannot be directly determined from the search phrase, the category can be indirectly determined based on the word included in the search phrase. The information search device 2 can determine the category more accurately when estimating the category by machine learning than when determining the category by non-machine learning.

  In other words, the information search device 2 according to the present embodiment further includes a category determination unit 23 and a search storage device 28. The category determination unit 23 determines the category of the search phrase. The search storage device 28 stores words corresponding to categories. The category determination unit 23 includes a category determination unit 231 and a category estimation unit 232. The category determination unit 231 determines a category based on whether or not a word included in the search phrase matches a word stored in the search storage device 28 in advance. The category estimation unit 232 estimates a category from words included in a search phrase that is not determined by the category determination unit 231 by machine learning.

  The information search apparatus 2 according to the present embodiment can indirectly estimate a category from words included in a search phrase by machine learning even if a word indicating a category directly is not included in the search phrase. By including the category determination unit 231 and the category estimation unit 232, the information search device 2 can determine a category with a lower processing load than a configuration in which a category is determined only by machine learning.

  The feature word generation unit 24 is configured to generate a feature word related to a word representing a search target. The feature word generation unit 24 generates a feature word from the search phrase transmitted from the information processing terminal 1. The feature word generation unit 24 performs morphological analysis on the search phrase in order to generate a feature word from the search phrase. For example, the feature word generation unit 24 acquires a word obtained by morphological analysis of the search phrase by the position information acquisition unit 22, the category determination unit 23, the search unit 25, or a morpheme analysis unit (not shown), and generates a feature word. It may be configured. The feature word generation unit 24 performs distributed expression in which the morpheme-analyzed words are expressed in numerical values. The feature word generation unit 24 summarizes the utilization forms of words subjected to morphological analysis. The “feature word” may be a single word or a group of a plurality of words. “Generating feature words” includes determining a feature word or selecting a feature word.

  For example, the feature word generation unit 24 replaces a word obtained by decomposing a search phrase by morphological analysis with a feature word having high similarity with the word by machine learning. The feature word generation unit 24 may replace one word with a feature word having a high degree of similarity among the words decomposed by the morphological analysis of the search phrase, or replace two or more words with those words having a similarity degree. It may be replaced with a high feature word. The feature word generation unit 24 generates a feature word having a higher similarity to the word representing the search target from the words included in the search phrase by machine learning. For example, the feature word generation unit 24 can select a desired feature word based on the similarity between the word and the feature word stored in advance after performing morphological analysis on the search phrase. More specifically, the feature word generation unit 24 can determine how far away the feature word is by measuring the feature vector of the feature word with cosine similarity. In other words, the feature word generation unit 24 can determine the similarity by measuring the feature vector of the feature word with cosine similarity.

  The feature vector can be obtained, for example, by an algorithm such as word2vec using, as input data, data obtained by numerically expressing a word subjected to morphological analysis. The algorithm such as word2vec used here is machine-learned so that the feature vector of the input word can be output. In cosine similarity, the cosine of two vectors is calculated, and the cosine value is used as the similarity. The cosine similarity represents the closeness of the angle formed by the vectors, and the closer to 1, the more similar. The cosine similarity indicates that the closer to 0, the less similar. The feature word generation unit 24 may calculate statistical data of search phrases according to the similarity. The feature word generation unit 24 is preferably configured to preferentially output a feature word having a higher frequency based on statistical data among a plurality of feature words.

  The feature word generation unit 24 extracts co-occurrence words from words included in the search phrase. In this case, the feature word generation unit 24 may use co-occurrence words as feature words, and the search unit 25 may search for a search target from the feature words. By using the co-occurrence words, the feature word generation unit 24 can effectively derive the search object even if it is a word that does not seem to be meaningful to express the search object and does not have a correlation. In other words, the feature word generation unit 24 generates a feature word, which is a feature vector that can more easily represent a search target, from the feature vector by machine learning.

  That is, the feature word generation unit 24 generates a feature word related to the word representing the search target from the search phrase. In the information search device 2 according to the present embodiment, the feature word generation unit 24 generates a feature word having a higher similarity to the word representing the search target from words including nouns, adjectives, verbs, and adverbs in the search phrase. To do.

  The information search device 2 can detect a search target with higher accuracy by generating a feature word having a higher degree of similarity to the word representing the search target. In particular, when the word included in the search phrase is an adjective, the information search device 2 generates a feature word from a word including at least the adjective, and therefore can search for a search target including a subjective or emotional meaning. .

  The search unit 25 is configured to be able to search for a search target from the accumulated database. In the accumulation database, feature words are classified into categories. The search target has point information indicating a specific point. The specific point represents the position of the search target on the map. The point information is, for example, longitude information and latitude information. The point information may include altitude information in addition to longitude information and latitude information. The search unit 25 acquires position information from the position information acquisition unit 22. The search unit 25 acquires category information from the category determination unit 23. The search unit 25 acquires a feature word from the feature word generation unit 24. The search unit 25 searches for a search target from the accumulation database based on the position information, the category, and the feature word. The search target searched by the search unit 25 corresponds to a feature word in a predetermined range and a determined category from the position indicated by the position information. In other words, the search unit 25 searches for a search object corresponding to the feature word in a predetermined range and a determined category from the position indicated by the position information, based on the position information, the category, and the feature word. It is configured to be able to.

  The timer unit 26 is configured to be able to measure the year, date, and time. The timer unit 26 may be configured so as to be able to measure time by counting the number of clocks of the CPU, for example. For example, the time measuring unit 26 can specify the date and time when the search communication device 21 received the search phrase from the information processing terminal 1. The timer 26 can discriminate the search phrase received by the search communication device 21 for each predetermined period. The timer 26 can store the date and time when the search phrase is received in the search storage device 28 in association with the statistical data.

  The search storage device 28 has a volatile storage device whose information disappears when the power is turned off and a non-volatile storage device whose information does not disappear even when the power is turned off. The volatile storage device temporarily stores information processed by the search control device 27. An example of a volatile storage device is a RAM.

  The nonvolatile storage device is a storage capable of storing a position information acquisition program, a category determination program, a feature word generation program, a search program, and a time measurement program. An example of the nonvolatile storage device is a ROM. The non-volatile storage device may store a place name database, a category database, statistical data, or a system dictionary.

  In the information search system 10 according to the present embodiment, the feature word generation unit 24 uses, for example, a “night view”, “beautiful”, “restaurant” using morphological analysis when the content of the search phrase is “beautiful restaurant with night view”. Get the word. The feature word generation unit 24 determines the similarity of the feature words co-occurring with “night view”, “beautiful”, or “restaurant” by machine learning, and the search unit 25 stores the accumulated database based on the determined feature words. The location information of “restaurant” whose search target is ambiguous can be searched.

  The information storage device 3 includes a storage control device 32 and a storage communication device 33 in addition to the storage device 31. The information storage device 3 is constituted by a database server, for example.

  The storage device 31 stores a storage database in which search targets to be searched by the information search device 2 are stored. In the accumulation database, search objects are classified for each category. In the storage database, search objects are stored in association with feature words.

  The storage device 31 has a volatile storage device whose information disappears when the power is turned off and a non-volatile storage device whose information does not disappear even when the power is turned off. The volatile storage device temporarily stores information processed by the accumulation control device 32. Examples of the volatile storage device include a RAM. The nonvolatile storage device is a storage capable of storing various information processing programs and various information. An example of the nonvolatile storage device is a ROM. The non-volatile storage device stores an accumulation database. The non-volatile storage device may store map information indicating a map and point information added to the map information.

  The storage control device 32 is electrically connected to the storage device 31 and the storage communication device 33. The storage control device 32 is configured to control the storage device 31 and the storage communication device 33. The accumulation control device 32 can be configured using a CPU or MPU. The accumulation control device 32 is configured to be able to implement a predetermined program based on a program stored in advance in the accumulation storage device 31, for example.

  The storage communication device 33 is configured so that the information search device 2 can access it. The storage communication device 33 is configured to receive information for searching for a search target from the information search device 2. The storage communication device 33 is configured to transmit information to be searched to the information search device 2.

  Further, the information storage device 3 may be configured such that a storage database can be constructed from an SNS (Social Networking Service) provided by an external information processing device (not shown). The information storage device 3 constructs a storage database from the generated feature words in the same manner as the information search device 2 generates a feature word from the search phrase based on the posted article information published externally by the external information processing device. can do. The posted article information is, for example, information on posted articles posted by the user to the external information processing apparatus.

  Next, the operation of the entire information search system 10 according to the present embodiment will be described.

  First, the relationship between the information processing terminal 1 and the information search device 2 will be described. In the information search system 10 according to the present embodiment, the user inputs a search phrase with the input device 11 of the information processing terminal 1. When the input device 11 receives an instruction to transmit a search phrase, the information processing terminal 1 transmits the search phrase from the terminal communication device 12 to the information search device 2.

  In addition, the information processing terminal 1 transmits position information indicating the selected position from the terminal communication device 12 to the information search device 2 when the user selects a position on the map displayed on the display device 13. It may be configured. In the information search device 2, when the position information from the information processing terminal 1 is received, the position information can be set as a starting point on the search map.

  Next, operation | movement of the information search device 2 which received the search phrase from the information processing terminal 1 is demonstrated using FIG.

  In the flowchart shown in FIG. 2, among the steps performed by the information search device 2, the information search device 2 receives a search phrase from the information processing terminal 1 (S 101). The information search device 2 stores the received search phrase in the search storage device 28.

  The information search device 2 performs a position information acquisition step of acquiring position information by a position information acquisition program whose operation is controlled by the search control device 27 (S102). The location information acquisition program acquires location information when location information is transmitted from the information processing terminal 1 in the location information acquisition step. When there is no position information that can be acquired in the position information acquisition step, the position information acquisition program acquires position information from the search phrase.

  Next, the information search device 2 performs a category determination step of determining the category of the search phrase by the category determination program whose operation is controlled by the search control device 27 (S103). The category determination program determines a category name from the category database when a word indicating a place name is included in the search phrase. When the search phrase does not include a word indicating the place name, the category determination program generates a search word from the search phrase by machine learning. The category discrimination program estimates a category name from the category database based on a search word having a high similarity among the generated search words.

  Subsequently, the information search device 2 performs a feature word generation step of generating a feature word related to the word representing the search target from the search phrase by the feature word generation program whose operation is controlled by the search control device 27 (S104).

  Next, the information search device 2 performs a search execution step of searching for a search target from the stored database by a search program whose operation is controlled by the search control device 27 (S105).

  Finally, the information search device 2 may be terminated if the search target is searched.

  According to the information search system 10 of the present embodiment, even if a user inputs a search phrase based on an ambiguous image to the information processing terminal 1, the information search device 2 performs a soft search using the user's subjective words or emotion words. It can be carried out. For example, the information search device 2 displays on the display device 13 of the information processing terminal 1 the location of a shop where the hot spicy ramen may be eaten on the display device 13 even if the search phrase “hot spicy ramen” is input. Can be displayed with a marker of a color corresponding to the category of ramen on the map.

  That is, in the information search system 10 of the present embodiment, the information processing terminal 1 further includes a terminal storage device 15. The terminal storage device 15 stores map information. The display device 13 displays a specific point indicating the search target on the map of the map information.

  The information search system 10 of the present embodiment can display a search target point with higher accuracy on a map.

  Moreover, in the information search system 10 of this embodiment, the feature word generation part 24 calculates the statistical data of a search phrase. The information storage device 3 associates the search phrase received by the search communication device 21 within the period set in advance from the information processing terminal 1 with the statistical data and stores it in the storage database.

  The information search system 10 of this embodiment can detect a search object with higher accuracy based on a search phrase within a preset period.

  Furthermore, in the information search system 10 of the present embodiment, the feature word generation unit 24 generates a feature word having a higher degree of similarity with the word representing the search target from the words included in the search phrase. The feature word generation unit 24 is configured to preferentially output feature words with higher frequency based on statistical data.

  The information search system 10 of this embodiment can detect a search target with higher accuracy.

  In the information search system 10 of the above-described embodiment, for example, in the case where the information processing terminal 1 has a touch panel, the map displayed on the display device 13 is enlarged or reduced, and the user searches as an area to be searched. By tapping a place to be designated, position information can be input to the input device 11.

  Further, the information processing terminal 1 displays the search target on the display device 13 so as to display the search target as a marker on the map. An example of the marker is a balloon in which a pictogram representing a category is displayed at the center. For example, the information processing terminal 1 may display a map including the marker on the display device 13 by changing the color or shade of the marker for each category. For example, the information processing terminal 1 may display the map including the marker on the display device 13 by changing the size of the marker according to the similarity between the word included in the search phrase and the feature word.

  Furthermore, in the information search system 10 of the above-described embodiment, the information processing terminal 1 may have a navigation function. Since the information processing terminal 1 has a navigation function, the user can be guided to the point information on the map displayed on the display device 13. The navigation function is stored in the terminal storage device 15 of the information processing terminal 1 and controlled by the terminal control device 14. The navigation function may be stored in the search storage device 28 of the information search device 2 and controlled by the search control device 27. The information search system 10 may be equipped with a navigation function in either the information processing terminal 1 or the information search device 2 as long as the map and navigation can be displayed on the display device 13 of the information processing terminal 1.

  Specifically, the navigation function may be stored in one of the information search device 2 and the information processing terminal 1, and the navigation function may be controlled by the other device. Alternatively, the navigation function may be divided and stored in both the information search device 2 and the information processing terminal 1, and either one device or both devices may cooperate to control the navigation function. Good.

  Each process or each function of the above embodiment may be realized by centralized processing by a single device or a single system, or may be distributedly processed by a plurality of devices or a plurality of systems. It may be realized with. In addition, each component in the above embodiment may be configured by dedicated hardware. Each component of the above embodiment may be realized by executing a program for components that can be realized by software.

  Each component of the above embodiment may be realized, for example, by a CPU executing a software program recorded on a recording medium. The program may be executed by being downloaded from a server, or may be executed by reading a program recorded on a predetermined recording medium. Further, the computer that executes the program may be singular or plural. The above embodiment may be configured to perform centralized processing or may be configured to perform distributed processing.

  It should be noted that the present invention can be implemented in various other forms without departing from the spirit, gist, or main features. The above embodiment is merely an example in all respects, and should not be interpreted in a limited manner. That is, taking the above embodiment as an example, the present invention is an information search apparatus or an information search system.

[Summary of Embodiment]
(1: About feature words)
According to the information search device 2 described above, it is assumed that the search phrase input by the user performing the search does not directly indicate the spot that the user wants to search due to the function of each unit included in the search control device 27. In addition, a reasonable spot can be detected and presented to the user.

  Specifically, the search control device 27 includes a feature word generation unit 24 and a search unit 25. Among these, the feature word generation unit 24 generates a feature word from a word obtained by morphological analysis of the search phrase. Hereinafter, a word obtained by morphological analysis of a search phrase may be referred to as an input word. In the present specification, the word “generation” is used in the meaning of “specific”, “determination”, “detection”, or “selection” with respect to the feature word.

  Here, the feature word generated by the feature word generation unit 24 may be a word having high similarity to the input word, may be a co-occurrence word of the input word, or may be both of them. Good. In the following, a word whose similarity to an input word is higher than a predetermined threshold may be called a similar word.

  When using a similar word as a feature word, the feature word generation unit 24 generates feature information indicating the feature of the input word. The feature information only needs to indicate the feature of the input word. For example, a feature vector of the input word may be used as the feature information. The feature vector can be calculated by using, for example, GloVe, WordNet, or fastText algorithm in addition to the above-described word2vec. Then, the feature word generation unit 24 calculates the similarity between the feature vector of each feature word associated with the spot in the accumulation database and the calculated feature vector, and identifies a word whose similarity is equal to or greater than a threshold value. . Thereby, the feature word production | generation part 24 can specify the similar word similar to the input word among the feature words matched with the spot in the accumulation database. The method for calculating the similarity is not particularly limited, and can be calculated by cosine similarity, for example.

  On the other hand, when using a co-occurrence word as a feature word, the feature word generation unit 24 specifies a co-occurrence word of the input word. The co-occurrence words can be extracted by, for example, TF-IDF (Term Frequency-Inverse Document Frequency). Further, the feature word generation unit 24 may use a co-occurrence word of the above-mentioned similar words as a feature word.

  And the search part 25 searches the accumulation | storage database memorize | stored in the accumulation | storage storage apparatus 31 using the produced | generated feature word, and detects the spot matched with the feature word in this accumulation | storage database. “Spot” means, for example, a place or place, or a building installed at a specific place or place. In the accumulation database, each spot is associated with position information indicating the geographical position of the spot. As described above, since the feature word used for the search is a similar word or a co-occurrence word of the input word, even if the search phrase does not directly indicate the spot that the user wants to search, a valid spot is selected. It can be detected and presented to the user.

  Note that the feature word generation unit 24 may use a word with a predetermined part-of-speech among the words obtained by morphological analysis of the search phrase. The predetermined part of speech may be, for example, a noun, an adjective, a verb, and an adverb. Thereby, it is difficult to generate a significant feature word, and for example, part of speech such as a conjunction, a moving verb, a particle, and an auxiliary verb can be excluded to generate a highly relevant feature word.

(2: About location information)
The search control device 27 may include the position information acquisition unit 22. The position information acquisition unit 22 acquires position information serving as a reference for determining a spot search range based on information received from the information processing terminal 1 used by the user. And the search part 25 sets a predetermined range on the basis of the position which the acquired positional information shows, and detects the spot in the range. This makes it possible to search for spots that are narrowed down to an area of interest to the user performing the search.

  The position information acquisition unit 22 may acquire the position information from a word obtained by morphological analysis of the search phrase. For example, if the word obtained by morphological analysis of the search phrase includes a word indicating the position, the position information acquisition unit 22 uses a place name database in which the word indicating the position is associated with the position information. Position information may be acquired. In addition, as a word which shows a position, a place name is mentioned, for example.

  Further, the position information acquisition unit 22 may acquire position information using a position information database for specifying position information. The position information database is a database in which a word and position information indicating a position corresponding to the word are associated with each other. When the position information acquisition unit 22 acquires position information using the position information database, the position information acquisition unit 22 uses at least one of a word obtained by morphological analysis of the search phrase, a similar word of the word, and a co-occurrence word of the word. To search the location information database. Thereby, the positional information acquisition part 22 can acquire the positional information relevant to the word obtained by carrying out the morphological analysis of the search phrase. Similar words and co-occurrence words may be specified using a machine-learned model. In this case, the position information acquisition unit 22 acquires position information estimated by machine learning from words included in the search phrase.

  In addition, the information processing terminal 1 may acquire position information indicating the position of the own apparatus using, for example, GPS (Global Positioning System), and may transmit the acquired position information to the information search device 2. In this case, the position information acquisition unit 22 may acquire the position information transmitted from the information processing terminal 1. By using such position information, it is possible to detect spots around the current location of the information processing terminal 1.

(3: About categories)
The search control device 27 may include a category determination unit 23. In addition, the category determination unit 23 may include a category determination unit 231 and a category estimation unit 232. The category discriminating unit 23 determines a category of words obtained by morphological analysis of the search phrase. In this case, the spot category is associated with each spot in the accumulation database. Thereby, the search part 25 becomes possible [detecting the spot of the category which the category discrimination | determination part 23 determined among the spots according to a search phrase].

  For example, the category determination unit 231 can determine a category corresponding to a word obtained by morphological analysis of a search phrase by using a category database in which a word is associated with a category corresponding to the word.

  The category estimation unit 232 uses a machine-learned algorithm to specify a category corresponding to a word obtained by morphological analysis of the search phrase. The algorithm may be, for example, machine-learned correspondence between word feature information and categories. The word feature information is, for example, a feature vector. When using an algorithm obtained by machine learning of the correspondence between word feature information and categories, the category estimation unit 232 obtains word feature information obtained by morphological analysis of the search phrase, and uses the obtained feature information as the algorithm. By inputting, a category can be specified. The category estimation unit 232 may specify at least one of a word similar word obtained by morphological analysis of the search phrase and a word co-occurrence word obtained by morphological analysis of the search phrase. Then, the category estimation unit 232 may specify a category by searching the category database using at least one of the specified similar word and co-occurrence word. Similar words and co-occurrence words may be specified using a machine-learned model. In this case, the category estimation unit 232 estimates the category from the words included in the search phrase by machine learning. For example, the category estimation unit 232 may calculate the similarity between a feature vector of a word belonging to a certain category and a word feature vector obtained by morphological analysis of the search phrase. In this case, if the calculated similarity is equal to or greater than the threshold, the category estimation unit 232 determines that the word obtained by morphological analysis of the search phrase belongs to the certain category.

  The determined category may be used for weighting described later, or may be used for narrowing the search range of the search unit 25. In the latter case, the search unit 25 associates the feature words generated by the feature word generation unit 24 with the spots belonging to the category determined by the category determination unit 23 among the spots stored in the storage database as search targets. Detecting a spot. For example, let us consider a case where the category determination unit 23 determines that the category of the word “landscape” included in the search phrase is “sightseeing spot”. In this case, the search unit 25 searches for spots associated with the category “sightseeing spot” among spots stored in the storage database, using feature words generated from “landscape”. Thereby, it becomes possible to narrow down a search range appropriately and to obtain a reasonable search result quickly.

(4: Priority in search)
When the feature word generation unit 24 can generate a plurality of feature words, the feature word generation unit 24 may generate a plurality of feature words with high priority. For example, in the information search system 10, the priority is calculated based on the cumulative number of searches used in the past, the frequency used in searches in the past predetermined period, and / or the cumulative count and usage frequency. It may be at least one of the following. For example, the feature word generation unit 24 can generate “scenery” and “restaurant” as feature words, and the search frequency of “restaurant” in the most recent week is higher than the search frequency of “scenery” in the most recent week. If it is high, “restaurant” may be used as a feature word.

  Note that the feature word generation unit 24 calculates information indicating the number of times and the frequency as described above for a search phrase received from the information processing terminal 1 within a preset period, and stores it as statistical data. That's fine. The stored statistical data is associated with the search phrase received from the information processing terminal 1.

  In addition, when the feature word generation unit 24 generates a plurality of feature words, the search unit 25 may perform a search using a plurality of feature words having high priority. Further, the search unit 25 may output the search result by the feature word having a higher priority in preference to the search result by the feature word having a lower priority. For example, the search unit 25 may sequentially output the search results using feature words with high priority. The statistical data can also be used for weighting described later.

(5: Construction of storage database)
The information storage device 3 can acquire various information related to the search target spot and construct a storage database. For example, the information storage device 3 may generate a feature word from an article posted to the SNS as described above, or an article posted on a website that introduces a spot or a site that includes spot review information. A feature word may be generated from Hereinafter, an article posted on the SNS or an article posted on a site posting word-of-mouth information on a spot is also referred to as a comment. Moreover, a website may be abbreviated as a site.

  For example, when the information storage device 3 obtains a comment “view is the best” posted on the word-of-mouth site about the spot “observation platform X”, the information storage device 3 performs a morphological analysis on the comment “view” and “best”. Get the word. Next, the information storage device 3 calculates feature vectors of these words, and uses the calculated feature vectors to generate feature words that are similar to each word. Then, the information storage device 3 adds the spot “observation platform X” to the storage database, and records the words obtained by the morphological analysis in association with the feature words generated from the words. If the spot “observation platform X” has already been recorded in the accumulation database, a word obtained by morphological analysis and a feature word generated from the word are additionally associated with each other and recorded. .

  Further, instead of the feature word or in addition to the feature word, a co-occurrence word or a cluster described later may be recorded in association with each other. In addition, on the sites as described above, the spot may be evaluated by the score or the number of stars. In that case, the information storage device 3 may also record the spot evaluation in the storage database. In addition, for example, the number of accesses to the site, the number of accesses to the page of the spot, the number of comments to the spot, the posting date and time of the comment, and the site name may be recorded in the accumulation database. The site name may be a service name, for example. Such information can be used, for example, for spot weighting, and for example, it is possible to preferentially search for spots with high evaluation.

  As described above, the information storage device 3 can expand the storage database based on the word-of-mouth information, thereby making it possible to search for new spots or based on the impressions of people who actually visited the spots. It is also possible to make a search possible. For example, if a feature word is recorded based on the comment “view is the best” for the spot “observation platform X”, the spot “observation platform X” can be identified from an ambiguous search phrase such as “the feeling of liberation is highest”. Can also be detected.

  In addition, by constructing an accumulation database based on information related to spots, it is possible to detect spots according to the user's will even when an ambiguous search phrase is input as in the above example. Is possible. On the other hand, in a general search site, since various objects other than spots must be included in the search object, a spot according to the user's intention with an ambiguous search phrase cannot usually be detected. For example, even if you search for the phrase “the feeling of liberation” on a general search site, the spot “observation platform X” is not detected, or even if it is detected, it is buried in other miscellaneous search results and difficult to recognize. It will become a state.

[Search using clusters]
As described in the above embodiment, the feature word generation unit 24 may set one word as a “feature word”, or may set a word group including a plurality of words as a “feature word”. In the following, the latter case will be described in more detail based on FIG.

  When a word group consisting of a plurality of words is used as a feature word, a feature word group consisting of a plurality of feature words having similar feature information indicating the feature of the word is associated with a spot to be searched in the accumulation database. . As described above, the feature information is, for example, a feature vector. Hereinafter, the feature word group is referred to as a cluster. When a search phrase is input, the feature word generation unit 24 determines a cluster corresponding to the word based on the feature information of the word included in the search phrase. Then, the search unit 25 detects spots associated with the clusters determined by the feature word generation unit 24 from the accumulation database.

  According to this configuration, even when a search phrase that is not directly associated with a spot is input, it is possible to detect a reasonable spot that matches the intention of the user who performs the search. For example, when a cluster A composed of similar feature words such as “landscape” and “landscape” is associated with a spot A in the accumulation database, a search phrase including the word “scenery” is input. To do. In this case, since “scenery” is similar to “scenery” and “scenery”, it is determined that “scenery” belongs to cluster A, and therefore spot A can be detected from the accumulation database. Note that whether or not the feature information is similar can be determined based on, for example, whether or not the similarity is equal to or greater than a predetermined threshold.

  In addition, when a word group composed of a plurality of words is used as a feature word, a category database in which clusters and categories composed of a plurality of feature words with similar feature information indicating the feature of the word are associated may be used. In this case, the category determination unit 23 detects, from the category database, the category associated with the cluster determined based on the feature information of the word included in the search phrase. According to this configuration, even when a search phrase that is not directly associated with a category is input, it is possible to determine an appropriate category in accordance with the intention of the user who performs the search.

  FIG. 3 is a flowchart illustrating an example of processing for searching for spots based on clusters. In the following, a specific example of spot search according to the flow of FIG. 3 will be described with reference to FIGS. FIG. 4 is a diagram showing a specific example in which clusters are determined from search phrases, FIG. 5 is a diagram showing an example of a data table stored in a category database, and FIG. 6 is stored in a spot database. It is a figure which shows an example of the data table which is. The category database and the spot database correspond to the above-described accumulation database. Also, the acquisition of the search phrase and position information (S201, S202) in FIG. 3 is the same as in FIG.

  In S203, the feature word generation unit 24 calculates a feature vector of each word obtained by morphological analysis of the search phrase. FIG. 4 shows an example in which four search phrases of “Scenery is beautiful” as Example 1, “Scenery is beautiful” as Example 2, “Scenery is beautiful” as Example 3, and “Fun Dating” as Example 4 are input. ing. For example, when the search phrase of Example 1 is input, in S203, “Scenery” and “Pretty” are excluded from the words obtained by performing morphological analysis on “Scenery is beautiful” except for the particle “Ga”. A feature vector is calculated. In the illustrated example, an n-dimensional feature vector is calculated. n may be a natural number, for example, n = 300.

  In S204, the feature word generation unit 24 determines a cluster of each word obtained by morphological analysis of the search phrase. Specifically, the feature word generation unit 24 inputs each of the feature vectors calculated in S203 to a predetermined clustering model, whereby each word obtained by morphological analysis of a cluster corresponding to the feature vector, that is, a search phrase. Get the output of the cluster. For example, in the example of FIG. 4, both “scenery” and “scenery” are determined to belong to the cluster having the cluster ID “15”. Further, both “beautiful” and “beautiful” are determined to belong to the cluster having the cluster ID “8”. The clustering model is a model that is machine-learned so as to output a cluster corresponding to a word using a feature vector of the word as input data. A cluster corresponding to a word is a cluster that includes a feature word having a feature vector similar to that word as a constituent element. A clustering model generation method will be described later.

  In S205, the category determination unit 23 determines the category of each word obtained by morphological analysis of the search phrase based on the cluster determined in S204. For this determination, a category database in which cluster IDs, feature words, weights, and categories are associated as shown in FIG. 5 can be used.

  Specifically, first, the category determination unit 23 extracts a category associated with the cluster determined in S204 from the category database. For example, in the example 1 “Scenery is beautiful” in FIG. 4, it is determined in S204 that the clusters are ID “15” and ID “8”. Then, the category “restaurant” is extracted.

  Next, the category determination unit 23 calculates the similarity between each word obtained by morphological analysis of the search phrase and each feature word of the category corresponding to the word, and multiplies the calculated similarity by a weight. For calculating the similarity, a fast text inter-word similarity estimation model may be used. Then, the calculated values are totaled for each category, and the extracted categories are ranked in descending order of the total value. Hereinafter, the total value totaled for each category is also referred to as a score.

  For example, in the case of Example 1 “Beautiful scenery” in FIG. 4, the word “landscape” obtained by morphological analysis of the search phrase and the characteristic word “landscape” registered in the category database for the category “sights / historic sites”. The similarity is calculated. The same applies to the word “beautiful”. Then, the calculated similarity is multiplied by the weight registered in the category database. For example, if the similarity between the word “landscape” and the feature word “landscape” is 0.8, and the similarity between the word “beautiful” and the feature word “beautiful” is 0.7, the category “sights / historic sites” Is calculated as follows.

About the word “scenery”: similarity “0.8” × weight “0.88” = 0.704
About the word “beautiful”: similarity “0.7” × weight “0.76” = 0.532
Score of category “Historic sites / historic sites”: 0.704 + 0.532 = 1.236
When a plurality of feature words with the cluster ID “15” are registered in the category database, the calculation of similarity for each feature word is performed by multiplying the weights in the same manner as described above. The same applies to the cluster ID “8”. Then, the scores obtained by multiplication are summed for each category to calculate the score, and the extracted categories are ranked in descending order of the score.

  In S206, the search unit 25 searches for spots based on the clusters determined in S204. For this search, a spot database in which cluster IDs, feature words, weights, and spot IDs are associated as shown in FIG. 6 can be used.

  Specifically, first, the search unit 25 extracts spots associated with the clusters determined in S204 from the spot database. For example, in Example 1 “Scenery is beautiful” in FIG. 4, the clusters are determined to be ID “15” and ID “8” in S204. “Restaurant” is extracted.

  Next, the search unit 25 calculates the similarity between each word obtained by morphological analysis of the search phrase and each feature word of the spot corresponding to the word, and multiplies the calculated similarity by a weight. For calculating the similarity, a fast text inter-word similarity estimation model may be used. Then, the calculated values are totaled for each spot, and the extracted spots are ranked in descending order of the total value. Hereinafter, the total value obtained by adding the values calculated as described above for each spot is also referred to as a score.

  For example, in the case of Example 1 “Scenery is beautiful” in FIG. 4, the word “landscape” obtained by morphological analysis of the search phrase and the feature word “ The similarity of “landscape” is calculated. Since the word “beautiful” matches the feature word “beautiful” registered in the spot database, the similarity is 1. Then, the calculated similarity is multiplied by the weight registered in the spot database. For example, if the similarity between the word “landscape” and the feature word “landscape” is 0.9, the score of the spot “sights A” is calculated as follows.

About the word “scenery”: similarity “0.9” × weight “0.53” = 0.477
About the word “beautiful”: similarity “1.0” × weight “0.46” = 0.46
The score of the spot “Sightseeing A”: 0.477 + 0.46 = 0.937
When a plurality of feature words with the cluster ID “15” are registered in the spot database, the calculation of the similarity similar to the above is performed for each feature word and multiplication by weight. The same applies to the cluster ID “8”. Then, the values obtained by multiplication are summed for each spot, and the extracted spots are ranked in descending order of the score which is the total value.

  In S207, the search unit 25 finalizes the spots detected in S206, taking into account the category determination result in S205. For example, in the category determination in S205, the search unit 25 may rank spots belonging to the category having the highest score in descending order of the score calculated in S206. Note that the category to which each spot belongs is determined in advance.

  Then, the search unit 25 ranks all the spots belonging to the category having the highest score, and then ranks all spots belonging to the category having the next highest score. It may be repeated until is determined. Note that the search unit 25 may end the repetitive processing when it is determined to a predetermined rank. As the predetermined order, for example, the top 20 is listed.

  The ranking method based on the category and spot scores is not limited to the above example. For example, for each spot detected in S206, the search unit 25 may calculate the total value of the score of the spot and the score of the category to which the spot belongs, and rank the spots in the order of the total value. In this case, it can be said that the category score is used for weighting the spot score. Further, the above-described statistical data may be added to the ranking in S207. In this case, for example, the ranking of spots with high evaluation can be made higher.

  Finally, in S209, the search communication device 21 notifies the information processing terminal 1 of the search result. As a result, the terminal control device 14 of the information processing terminal 1 displays the search result on the display device 13. At this time, the search communication device 21 may notify the information processing terminal 1 of only a predetermined number of higher ranks determined in S208 among the search results. As the upper predetermined number, for example, the top 10 of the search results can be cited.

  The search communication device 21 may also notify the information processing terminal 1 of the order determined in S208. In this case, the terminal control device 14 can display the search results on the display device 13 in a display mode according to the order. For example, when displaying the search result on a map, a larger marker may be displayed on a spot having a higher rank on the map. Thereby, it is possible to make the user recognize which of the detected spots is more appropriate.

  As described above, it is possible to detect appropriate spots from various search phrases by determining clusters corresponding to the search phrases and searching for spots using the determined clusters. For example, even if the search phrase is Example 1 “Scenery is beautiful” in FIG. 4, Example 2 “Scenery is beautiful” in FIG. 4, and Example 3 “Scenery is beautiful” in FIG. Spots can be detected. In addition, when the search phrase is Example 4 “fun date” in FIG. 4 which is completely different from Examples 1 to 3, a cluster different from Examples 1 to 3 is determined, and as a different spot accordingly, for example, amusement park F Is detected.

  The cluster can also be used to specify a category. By specifying a category, spots in the category can be detected. Further, by setting a higher weight for a combination of a spot and a feature word or a combination of a category and a feature word for a combination having higher relevance, it is possible to preferentially detect a spot with high validity.

(Modification of category judgment / spot search)
In the category determination of S205, the similarity between the word obtained by morphological analysis of the search phrase and the feature word registered in the category database is calculated, but this process can be omitted. In this case, in S205, the category determination unit 23 may calculate the sum of the weight values of each cluster ID associated with the category, and use the calculated value as the score of the category.

  Similarly, the calculation of similarity can be omitted in the spot search in S206. In this case, in S206, the search unit 25 may calculate the sum of the weight values of each cluster ID associated with the spot, and use the calculated value as the score of the category.

[Generation of category database and spot database]
(1: Configuration to achieve)
The information search system 10 may have a category database and a spot database generation function. The components for realizing this function may be provided in the information search device 2 or the information storage device 3, for example. Further, a device having this function may be added to the information search system 10 separately.

  Below, the example in which the information storage device 3 is provided with the production | generation function of a category database and a spot database is demonstrated based on FIG. FIG. 7 is a block diagram illustrating an example of a main configuration of the information storage device. As illustrated, the accumulation control device 32 includes a clustering model generation unit 321, a feature word extraction unit 322, a cluster determination unit 323, a spot database generation unit 324, a category classification unit 325, and a category database generation unit 326. Hereinafter, the database may be abbreviated as DB. The functions of these units can also be realized by a program, for example. In addition, it is assumed that word-of-mouth data written in a predetermined site is used to generate the category database and the spot database. Further, it is assumed that at least one sentence is included in one word of mouth. Here, the predetermined site may be a plurality of sites or an SNS. In the following, word-of-mouth data may be simply referred to as word-of-mouth.

  The clustering model generation unit 321 generates a clustering model for clustering words. Clustering may be performed based on, for example, the similarity of feature vectors, whereby roughly similar word groups can be handled as a cluster. The “roughly similar word group” can be said to be a rough synonym group. For example, a fastText vectorization model may be used to calculate a feature vector of a word. The generated clustering model is used for determining a cluster in S204 of FIG. By performing clustering, a search can be performed in units of clusters, so that the search can be made efficient and the search can be speeded up.

  The feature word extraction unit 322 extracts feature words in the input sentence. For example, co-occurrence word analysis can be used to extract feature words. Specifically, the feature word extraction unit 322 calculates a TF-IDF value for each word included in the word group obtained by performing morphological analysis on the input sentence. Then, the feature word extraction unit 322 ranks words included in the word group according to the calculated TF-IDF value, and extracts words ranked higher as feature words. Further, since the TF-IDF value is a numerical value indicating the probability of the feature word, the feature word extraction unit 322 sets the TF-IDF value of the extracted feature word as the weight value of the feature word.

  The cluster determination unit 323 determines a cluster of feature words extracted by the feature word extraction unit 322 using the clustering model generated by the clustering model generation unit. The determined cluster includes a plurality of feature words having similar feature information as components.

  The spot DB generation unit 324 generates a spot database as shown in FIG. 6 by associating spots, feature words, clusters thereof, and weight values. The feature word and the weight value are extracted and calculated by the feature word extraction unit 322. The cluster is determined by the cluster determination unit 323.

  The category classification unit 325 classifies the input sentence into a plurality of predetermined categories. For classification, a classification model constructed by supervised machine learning may be used. As the classification model, for example, a fast text text classification model may be used. In the machine learning, a sentence with a known category is used as teacher data. For example, for each word of mouth written on a predetermined site, a category may be associated with the teacher data manually.

  The category DB generation unit 326 generates a category database as shown in FIG. 5 by associating categories, feature words, clusters thereof, and weight values. The feature word and the weight value are extracted and calculated by the feature word extraction unit 322. The cluster is determined by the cluster determination unit 323.

(2: Process flow)
FIG. 8 shows an example of processing for generating a spot database. In S301, the clustering model generation unit 321 generates a clustering model. For example, the clustering model generation unit may calculate feature vectors for all words obtained by morphological analysis of all the reviews written on a predetermined site, and generate a clustering model based on the calculated feature vectors. . In this case, the generated clustering model is a model that classifies input words into clusters having similar feature vectors to the words. For example, fastText can be used to generate such a model.

  In S302, the feature word extraction unit 322 extracts feature words and weights from the word of mouth of each spot. Extraction of feature words and weights can be performed using the TF-IDF value as described above. The correspondence between each word of mouth and the spot is specified in advance according to which spot the word of mouth is written for. For example, if the word of mouth is written on a page introducing the famous place A in a predetermined site, the feature word extracting unit 322 stores the word of mouth in association with the famous place A so that the word of mouth is the word of mouth of the famous place A. Can be identified.

  In S303, the cluster determination unit 323 determines a cluster of feature words extracted in S302 using the clustering model generated in S301. A feature word, a weight, and a cluster are determined for each word of mouth written for each spot by the processing in S302 and the processing in S303.

  In S304, the spot DB generation unit 324 stores the feature words, weights, and clusters determined as described above in the spot database in association with the corresponding spots. Thereby, for each spot, a record in which the feature word, the cluster, and the weight are associated is added to the spot database. If the spot database has not been generated, a spot database is generated in S304.

  Next, FIG. 9 will be described. FIG. 9 is a flowchart illustrating an example of processing for generating a category database. In S401, the category classification unit 325 classifies each word of mouth written on a predetermined site into a category.

  In S402, the feature word extraction unit 322 extracts feature words and weights from the reviews of each category. Extraction of feature words and weights can be performed using the TF-IDF value as described above. In S402, the feature words and weights extracted in S302 may be used.

  In S403, the cluster determination unit 323 determines a cluster of feature words extracted in S402 using the clustering model generated in S301. By the process of S302 and the process of S403, a feature word, a weight, and a cluster are determined for each category of word-of-mouth.

  In S404, the category DB generation unit 326 stores the feature words, weights, and clusters determined as described above in the category database in association with the corresponding categories. Thereby, for each category, a record in which a feature word, a cluster, and a weight are associated is added to the category database. If the category database has not been generated, the category database is generated in S404.

(3: Summary)
As described above, the clustering model generation unit 321 calculates a feature vector for each word obtained by morphological analysis of the word of mouth, and generates a clustering model based on the calculated feature vector. According to the clustering model generated in this manner, search phrases including expressions similar to word-of-mouth can be clustered with high accuracy. In addition, this clustering model can be generated only from word of mouth.

  A spot database and a category database can be generated by using the built clustering model and word-of-mouth. The spot database and the category database can be generated only from reviews. Moreover, these databases can be expanded by acquiring the latest reviews as needed. The word of mouth reflects the latest terms used when evaluating or expressing spots and categories, so by using each database generated as described above, such latest terms can be used. It is possible to cope with the used search.

  The cluster determination unit 323 determines a cluster of feature words using the clustering model, and the spot DB generation unit 324 stores the clusters and spots determined by the cluster determination unit 323 in association with each other. Thereby, even if the search part 25 is a case where the search phrase which consists of the word which is not matched directly with a spot is input, it can detect a suitable spot by using the cluster corresponding to the word. It becomes possible.

  Similarly, the category DB generation unit 326 stores the clusters determined by the cluster determination unit 323 and the categories classified by the category classification unit 325 in association with each other. Thereby, even if the category discrimination | determination part 23 is a case where the search phrase which consists of the word which is not matched with the category directly is input, it determines a suitable category by using the cluster corresponding to the word. Is possible.

  Further, the feature word extraction unit 322 extracts feature words and calculates their weights, and the spot DB generation unit 324 stores spots, clusters, feature words, and weights in association with each other. As a result, for each feature word included in one cluster, the magnitude of the relationship with the spot can be specified by the weight value. Therefore, the search unit 25 can rank the search results in consideration of the degree of relevance between the feature word corresponding to the search phrase and the spot.

  Similarly, the category DB generation unit 326 stores categories, clusters, feature words, and weights in association with each other. As a result, for each feature word associated with one category, the magnitude of relevance with the category can be specified by the weight value. Therefore, the category determination unit 23 can rank the category determination results in consideration of the degree of relevance between the feature word corresponding to the search phrase and the category.

(About relearning)
When using a model generated by supervised machine learning in the search, teacher data corresponding to the search result of the search may be generated, and the model may be updated by performing machine learning again with the teacher data. For example, it is assumed that a certain search phrase is input, category determination (S204) and spot search (S206) are performed, and one of the spots notified to the information processing terminal 1 is selected by the user. In this case, it is considered that the selected spot was a spot along the user's will. Therefore, the search phrase or a word constituting the search phrase, its feature vector, and the spot or its category can be associated with each other as teacher data. For example, it is good also as teacher data for updating the classification model which a category classification | category part uses by matching the said search phrase and the category of the selected spot. By updating the classification model by re-learning using such teacher data, the classification accuracy in S401 can be improved.

DESCRIPTION OF SYMBOLS 1 Information processing terminal 2 Information retrieval apparatus 3 Information storage apparatus 10 Information retrieval system 11 Input apparatus 12 Terminal communication apparatus 13 Display apparatus 15 Terminal storage apparatus 21 Search communication apparatus 22 Location information acquisition part 23 Category discrimination | determination part 24 Feature word generation part 25 Search 28 Search and storage device 31 Storage and storage device

Claims (9)

A position information acquisition unit for acquiring position information indicating a position on the map as a starting point of the search;
A feature word generation unit that generates a feature word related to a word representing a search target from a search phrase;
A search unit that searches the search target from a storage database that stores the feature word and the search target corresponding to the feature word;
The search target has point information indicating a specific point,
The search unit searches for the search object corresponding to the feature word within a predetermined range from the position based on the position information and the feature word.   The information search device according to claim 1, wherein the position information acquisition unit acquires the position information estimated from words included in the search phrase by machine learning. A category discriminating unit for discriminating a category of the search phrase;
A search storage device storing words corresponding to the category;
In the accumulation database, the feature words are classified for each category,
The category discriminating unit
A category determination unit for determining the category based on whether or not a word included in the search phrase matches a word stored in the search storage device in advance;
A category estimation unit that estimates the category from words included in the search phrase that has not been determined by the category determination unit by machine learning;
The search unit searches the search target corresponding to the feature word in the predetermined range and the determined category based on the position information, the category, and the feature word. Or the information search device of Claim 2.   The feature word generation unit generates a feature word having a higher similarity to the word representing the search target from words including nouns, adjectives, verbs, and adverbs in the search phrase. The information search device according to claim 1. An information search system comprising an information processing terminal, an information search device capable of communicating with the information processing terminal, and an information storage device accessible by the information search device,
The information processing terminal
An input device for inputting a search phrase;
A terminal communication device that transmits the search phrase to the information search device and receives a search object searched based on the search phrase from the information search device;
A display device for displaying the search object,
The information search device includes:
A search communication device that receives the search phrase from the information processing terminal and transmits the search target;
A position information acquisition unit for acquiring position information indicating a position on the map as a starting point of the search;
A feature word generation unit that generates a feature word related to a word representing a search target from the search phrase;
A search unit for searching the search target from a storage database,
The search target has point information indicating a specific point,
The search unit searches the search object corresponding to the feature word within a predetermined range from the position based on the position information and the feature word,
The information storage device includes a storage device for storing the storage database,
The storage database stores the feature word and the search target corresponding to the feature word. The information processing terminal further includes a terminal storage device in which map information is stored,
The information display system according to claim 5, wherein the display device displays the specific point indicating the search target on a map of the map information. The feature word generation unit calculates statistical data of the search phrase,
7. The information storage device according to claim 5 or 6, wherein the search phrase received by the search communication device and the statistical data are associated with each other and stored in the storage database within a preset period from the information processing terminal. Information retrieval system described.   The feature word generation unit generates the feature word having a higher similarity to the word representing the search target from the words included in the search phrase, and the feature word having a higher frequency based on the statistical data. The information retrieval system according to claim 7, wherein the information is output with priority. In the accumulation database, a feature word group consisting of a plurality of feature words having similar feature information indicating a feature of a word is associated with the search target,
The feature word generation unit determines the feature word group corresponding to the word based on the feature information of the word included in the search phrase,
5. The search unit according to claim 1, wherein the search unit detects the search target associated with the feature word group determined by the feature word generation unit from the accumulation database. The information search device according to item.

Images