JP6803998B1 - Information processing device and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device and an information processing method for proposing contents suitable for a broadcasting time zone of an advertisement produced in the future. An information processing device and an information processing method of the present invention acquire input information including information related to a broadcast time zone of an advertisement to be produced in the future, and based on the input information, correspond to the broadcast time zone. An image element is generated, and the generated image element outputs image information representing an image of an advertisement to be produced in the future. [Selection diagram] Fig. 4

Description

The present invention relates to an information processing device and an information processing method, and more particularly to an information processing device and an information processing method for executing processing related to advertisements produced in the future.

Technology that supports transactions between the provider of the time zone (that is, the ad space) when the advertisement is broadcast and the advertiser, the advertising agency, etc. who advertise using the ad space has already been developed. .. One example thereof is the technique described in Patent Document 1.

The technique described in Patent Document 1 relates to a matching service for advertising space, and allows an advertiser to input the layer of people who want to be in contact with the advertisement (so-called target group), the advertising budget, etc., and the most suitable advertisement for the advertiser. Search the frame and display the search results on the advertiser's terminal. If the matching service described in Patent Document 1 is used, an advertisement space that matches the advertisement scheduled to be broadcast is searched, and the advertiser can perform a highly appealing advertisement by the searched advertisement space.

Japanese Unexamined Patent Publication No. 2008-14277

As another viewpoint for performing highly appealing advertisements, it is conceivable to produce advertisements with contents suitable for the inventory. That is, in order for a person who has acquired (purchased) a certain ad space to effectively use the ad space, the content of the advertisement that matches the program or the like broadcast within the time zone of the ad space is determined, and the content of the advertisement is determined. Creating an ad is important.

The present invention has been made in view of the above circumstances, and an object of the present invention is to solve the following object. That is, the present invention solves the above-mentioned problems, and specifically provides an information processing device and an information processing method for proposing contents suitable for the broadcasting time zone of an advertisement produced in the future. The purpose.

In order to achieve the above object, the data processing device of the present invention is based on an input unit that acquires input information including program information about a program to be broadcast in the broadcast time zone of an advertisement produced in the future, and an input unit. An acquired unit that generates an image element according to the broadcast time zone, an output unit that outputs image information representing an image of an advertisement to be produced in the future by the image element, and an acquired image acquired for each of a plurality of programs. A storage unit that stores program information and acquired advertisement information acquired for each of a plurality of advertisements broadcast in the past, and an acquired program that is stored in the storage unit by identifying a specific program based on input information. Based on the program information and the acquired advertisement information, the extraction unit that extracts the acquired advertisement information of the advertisement that matches the specific program, the words that frequently appear between the extracted acquired advertisement information, the words that are conceptually similar, And a learning unit that learns the tendency of words with a high degree of co-occurrence with these words, and the generation unit broadcasts from the learned tendency and input information during the broadcast time of the advertisement to be produced in the future. It is characterized in that a characteristic word that matches the program to be played is generated as an image element .

According to the information processing device configured as described above, an image element corresponding to the broadcast time zone is generated from the input information related to the broadcast time zone of the advertisement to be produced in the future, and the information by the image element is used. Output image information of advertisements produced in the future. As a result, it is possible to propose an image suitable for the broadcasting time zone of the advertisement to be produced in the future and support the production of the advertisement.

Further, in one embodiment of the present invention, the input unit uses program information related to a program to be broadcast during the broadcast time zone of an advertisement produced in the future and a product related to a type of product or service advertised in the advertisement produced in the future. It is advisable to acquire input information including such information, keywords related to the contents of advertisements to be produced in the future, and meta information of programs to be broadcast in the future.
If the above-mentioned information is acquired as input information, more accurate image elements can be generated based on the above-mentioned input information for the image of the advertisement to be produced in the future.

Further, in the above configuration, it is more preferable that the program information includes evaluation information regarding the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future. In this case, for example, an image element corresponding to the program can be derived based on the evaluation (specifically, the context described later) of the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future. As a result, it is possible to propose an image of an advertisement produced in the future that matches the evaluation of the program broadcast during the broadcast time of the advertisement.

Further, in one embodiment of the present invention, the generation unit may generate a plurality of words including a keyword as an image element. By generating a plurality of words including keywords as image elements in this way, it is possible to propose a more specific image for an advertisement to be produced in the future. Further, since the input keyword is included in the image element, a word group including at least a word (keyword) specified by the user can be proposed as representing an image of an advertisement to be produced in the future.

Further, in an embodiment of the present invention includes a raw made the image element, the verification unit for verifying the matching degree between program broadcast during the broadcast time period of ads produced future, validated degree of matching When the setting condition is satisfied, the output unit may output the image information including the generated image element.
According to the above configuration, the matching degree between the image element generated for the advertisement to be produced in the future and the program broadcast in the broadcasting time zone of the advertisement is verified, and when the matching degree satisfies the setting condition, the above Outputs image information including image elements. As a result, the image information of the advertisement to be produced in the future is proposed after confirming its validity, so that the credibility of the proposed image information is improved.

Further, in the above configuration, it is more preferable to have an image element in which the matching degree verified by the verification unit satisfies the setting condition and a calculation unit for calculating the matching degree with each of a plurality of programs to be broadcast in the future. ..
According to the above configuration, the degree of matching between the image of the advertisement produced in the future and the program to be broadcast in the future is calculated. By using the calculated matching degree, it is possible to grasp whether the broadcast time zone of each of a plurality of programs to be broadcast in the future is a time zone suitable for broadcasting the advertisement to be produced in the future. Can be done.

Further, in order to solve the above-mentioned problems, in the information processing method of the present invention, the computer acquires input information including program information about a program to be broadcast in the broadcast time zone of an advertisement to be produced in the future, and uses the input information as input information. Based on this, an image element according to the broadcast time zone is generated, and the image information representing the image of the advertisement to be produced in the future is output by the image element, and the acquired program information acquired for each of a plurality of programs and the past The acquired advertisement information acquired for each of the plurality of advertisements broadcasted in the above is stored in the storage unit, the specific program to be specified based on the input information is specified, and the acquired program information and the acquired program information stored in the storage unit are stored. Based on the advertisement information, the acquired advertisement information of the advertisement that matches the specific program is extracted, and the words that frequently appear among the extracted acquired advertisement information, the words that are conceptually similar, and the coexistence with these words. Learning the tendency of high-frequency words, and generating characteristic words that match the programs that will be broadcast during the broadcast time of future advertisements as image elements from the learned trends and input information. It is a feature.
According to the above method, it is possible to propose an image suitable for the broadcasting time zone of an advertisement to be produced in the future and support the production of the advertisement.

According to the present invention, it is possible to propose an image suitable for the broadcasting time zone of an advertisement produced in the future. As a result, it is possible to support the production of advertisements, for example, it is possible to produce highly appealing advertisements in consideration of the broadcasting time zone, and the person who purchased the advertisement space of the advertisement can use the advertisement space. It can be used effectively.

It is explanatory drawing of the service using the information processing apparatus of one Embodiment of this invention, and shows the provider of a service, and the person concerned. It is explanatory drawing of the service using the information processing apparatus of one Embodiment of this invention, and shows an example of the screen output to the user of the service. It is a figure which shows the structure of the service providing system including the information processing apparatus of one Embodiment of this invention. It is a block diagram which shows the functional part which the information processing apparatus of one Embodiment of this invention has. It is a figure which shows the output example of image information. It is a figure of the information processing flow of one Embodiment of this invention (the 1). It is a figure of the information processing flow of one Embodiment of this invention (the 2).

An information processing apparatus and an information processing method according to an embodiment of the present invention (the present embodiment) will be described in detail below with reference to the accompanying drawings.
It should be noted that the embodiments described below are merely examples for facilitating the understanding of the present invention, and do not limit the present invention. That is, the present invention may be modified or improved from the embodiments described below without departing from the spirit of the present invention. Also, of course, the present invention includes an equivalent thereof.

Further, in the present specification, the term "device" refers to one device that independently exerts a specific function, or a plurality of devices that exist in a distributed manner but cooperate to exert a specific function. Also includes.

Further, in the present specification, "broadcasting" means sending audio or audio and video by radio waves for the purpose of auditioning or listening to the public, or distributing digitized information via a network such as the Internet. Specifically, it corresponds to the broadcasting of television (including Internet television) or radio (including IP simul radio).

Further, in the present specification, the "program" is a unit of organization in broadcasting such as television or radio, and a time zone (date and time) and a broadcasting station to be broadcast are set in advance. The "broadcast time zone of a program" includes the time zone from the broadcast start time to the broadcast end time of the program, the time zone from a few minutes before the broadcast start to the broadcast start, and / or the broadcast from the broadcast end. It can also include time zones up to a few minutes after the end.

Further, in the present specification, the "advertisement" is a content, that is, a commercial, which is broadcast for the purpose of promoting information about a product, service, business, etc. (hereinafter, product, etc.) to a person for commercial purposes. It will be broadcast within a predetermined broadcasting time zone (advertising space). The "person" in the present specification may be an individual, or may be a group or group (for example, a household) to which a plurality of individuals belong.

<< About the service using the information processing device of this embodiment >>
A service using the information processing device of the present embodiment (hereinafter, this service) will be described with reference to FIGS. 1 and 2. 1 and 2 are explanatory views of this service, FIG. 1 shows a provider of this service and related persons, and FIG. 2 is a screen output to a user of this service. An example is shown.
The screens shown in FIGS. 2 and 5 described later are merely examples, and the screen configuration, the content of the displayed information, the GUI (Graphical User Interface), etc. are based on the system design specifications and user preferences. It can be freely decided according to the situation and can be changed as appropriate.

In the following, it is assumed that the broadcasting is television broadcasting (specifically, private television broadcasting), and in order to simplify the explanation, the number of private television broadcasting stations is one, although it is different from the actual one. I will do it. However, needless to say, the present invention is also applicable to radio broadcasting, and is naturally applicable to a case where the number of television broadcasting stations is plural.

This service proposes the content suitable for the broadcasting time zone of the advertisement to be produced in the future, and is provided from the service provider S to the service user T as shown in FIG. The "advertisement to be produced in the future" corresponds to, for example, an advertisement in which the product or the like to be advertised in the advertisement is determined, but the image or the like is undecided. The image of the advertisement means the content of the advertisement (specifically, the outline, the scene / scene, the performers and the characters appearing, etc.), the impression when the advertisement is viewed (listened), and the concept and concept of the advertisement.

As shown in FIG. 1, the service provider S acquires information (data) related to programs and advertisements from TV broadcasting station J, and information (data) related to evaluation of programs and advertisements from research company K, respectively. The service provider S creates a database of acquired information, accumulates it, and uses the accumulated information to provide this service.

The service user T is, for example, an advertiser, an advertising agency, an advertising production company, or the like, and uses this service through his / her own terminal. When using this service, the service user T inputs information about an advertisement to be produced in the future, and for example, regarding the broadcast time zone of the advertisement, the meta information of the program broadcast at the same time zone, and the above advertisement. Enter keywords (words to characterize the advertisement), etc.

When the input is completed, the service user T derives the image of the advertisement to be produced in the future from the input information. After that, the data related to the derived advertisement image is sent, and the data is expanded on the terminal of the service user, so that the screen shown in FIG. 2 is displayed. Information representing the image of the advertisement to be produced in the future (hereinafter referred to as image information) is displayed on this screen, and the service user T sees the information displayed on the screen and is a reference when producing the above advertisement. Can be. In the screen example of FIG. 2, as the image information of the advertisement to be produced in the future, a word group representing an image suitable for the broadcast time zone of the advertisement (specifically, "male", "drive", and "slope"). ) Is displayed.

Further, in the present embodiment, it is possible to grasp the future program (television program) that matches the image information together with the image information of the advertisement to be produced in the future. More specifically, on the screen shown in FIG. 2, the degree of matching between the image information and each of the plurality of programs to be broadcast in the future is displayed below the area where the image information is displayed. The plurality of programs to be broadcast in the future are programs to be broadcast in a predetermined period in the future set in advance. For example, they are broadcast on the television broadcasting station J on each day of the week including the broadcast date of the advertisement to be produced in the future. All programs to be broadcast are applicable.

The service user T finds a program having a high degree of matching with the image information displayed on the screen from a plurality of programs to be broadcast in the future, and broadcasts an advertisement based on the above image information during the broadcast time of the program. As a result, it is possible to carry out highly appealing advertisements.

The degree of matching is the value obtained by quantifying (vectorizing) the program information including the program name and the genre of the program and the image information of the advertisement to be produced in the future for each of the plurality of programs to be broadcast in the future. It can be calculated using (vector). As a method for calculating the degree of matching, a known method can be used. For example, the correlation coefficient may be obtained as the degree of matching, and the distance indicating the degree of similarity between the information (specifically, the Euclidean distance and the Mahalanobis distance). Alternatively, the cosine distance, etc.) may be calculated, or the absolute error between the information may be calculated. Further, as a method for digitizing (vectorizing) information, known methods such as word2vec and LDA (Latent Dirichlet Allocation) can be used.

<< About the configuration of the information processing device according to this embodiment >>
Next, the configuration of the information processing device (hereinafter, information processing device 10) according to the present embodiment will be described with reference to FIG. FIG. 3 is a diagram showing a configuration of a service providing system 100 including an information processing device 10.

The information processing device 10 uses various information related to programs and advertisements to perform a series of information processing related to this service. In the present embodiment, the information processing device 10 is composed of a server computer 11 managed by the service provider S and a storage device 12 connected to the server computer 11.

The server computer 11 corresponds to a computer that forms the main body of the information processing device 10, and performs a series of data processing related to this service. The server computer 11 may be an ASP (Application Service Provider) server, and in that case, a series of information processing related to this service is executed as an ASP service.

As shown in FIG. 3, the server computer 11 includes a processor 11a composed of a CPU and the like, a memory 11b composed of a ROM and a RAM, a communication interface 11c composed of a network card and the like, and a storage 11d composed of an HDD or SSD. It has an input device 11e including a mouse, a keyboard, and the like, and an output device 11f including a display, a speaker, and the like.

Further, an information processing program for this service (hereinafter, service program) is installed in the server computer 11. When the service program is read and executed by the processor 11a, the server computer 11 exerts a function as an information processing device 10 (details, the functions of each functional unit described later are exerted).

The storage device 12 corresponds to a storage unit, stores various types of information provided for this service, and accumulates the information to form a database. Specifically, the storage device 12 includes acquired program information acquired for each of the plurality of programs broadcasted in the past, and acquired advertisement information acquired for each of the plurality of advertisements broadcasted in the past. Remember. The acquired program information and the acquired advertisement information are each stored in the storage device 12 and stored in a database.

The acquired program information includes the metadata (metadata) of the program broadcasted in the past and the information of the viewer's evaluation regarding the program. Meta information is information including broadcast date and time, broadcasting station, broadcasting time (minutes), program name, program genre, whether it is a regular program or a special program, performers, contents, program sponsors, etc., for example, TV broadcasting. It is available from station J.
The meta information may include items other than the above, and may include, for example, an average audience rating and the like. Further, the source of the meta information is not limited to the television broadcasting station J, and may be obtained from other organizations such as a program production company. Further, the acquired program information is not limited to the information of the program broadcasted in the past, and may include the information of the pre-broadcast program to be broadcast in the future.

The evaluation information is obtained from the results of a questionnaire survey conducted by the research company K using the viewer as a monitor, and the impression, viewing feeling, and psychology of the viewer when the viewer actually watched the program. Represents a target state, the so-called context. The context includes impressions and emotions about the program, mood and feelings when watching the program, emotional desires expected from the program, emotions toward the performers of the program, and more specifically, favorability (favorability). , Familiarity, harassment, excitement, excitement, uplifting (fun mood), freshness, usefulness, informativeness, spectacularness, topicality, funny content, casual content, and boredom. Can be mentioned. The evaluation information can be obtained from, for example, the research company K, but is not limited to this, and may be obtained directly from the viewer (viewing household), for example.

In addition, the evaluation information is not limited to the context, and is a score indicating the degree of impressions and impressions corresponding to the context, such as "exciting 3.0 points" and "laughable 1.8 points". (Hereinafter, an evaluation index) may be used as evaluation information. The evaluation index can be obtained by a known technique for quantifying impressions and impressions, for example, LDA.

The acquired advertisement information includes the metadata (metadata) of the advertisement broadcasted in the past and the information of the viewer's evaluation regarding the advertisement. Meta information is information including broadcasting time (seconds), type (category) of products, names of products, providers (sales companies) of products, performers, contents, etc., for example, television broadcasting. It is available from station J.
The meta information may include items other than the above, and may include, for example, the number of broadcasts. Further, the source of the meta information is not limited to the television broadcasting station J, and may be obtained from other institutions such as an advertising production company or an advertising agency.

The evaluation information regarding the advertisement is obtained from the response result of the questionnaire survey conducted by the research company K using the viewer as a monitor, and represents the above-mentioned context, like the evaluation information regarding the program. The evaluation information can be obtained from, for example, the research company K, but is not limited to this, and may be obtained directly from, for example, a viewer (viewing household).

Further, the storage device 12 stores the transaction unit for the broadcast time zone of the advertisement, that is, the advertisement space information regarding the advertisement space, and stores the advertisement space in a database. The ad space information is information for specifying the ad space of the pre-broadcast advertisement to be broadcast in the future, and specifically, the date and time, the day, the number of seconds, and the same broadcast time zone of the ad space are broadcast. Information on programs can be mentioned. The advertising space information may be obtained from, for example, TV broadcasting station J, but may also be obtained from other institutions, such as a business company that handles advertising space.

Further, the storage device 12 stores program information related to pre-broadcast programs to be broadcast in the future, and stores the program information to form a database. The program information of the pre-broadcast program includes the broadcast date and time of the pre-broadcast program, the broadcasting station, the broadcasting time (minutes), the program name, the program genre, whether it is a regular program or a special program, the performers, and other meta information. , For example, it is available from TV broadcasting station J.
The program information of the pre-broadcast program may include items other than the above, and may include, for example, an expected audience rating. Further, the source of the program information of the pre-broadcast program is not limited to the television broadcasting station J, and may be obtained from other organizations such as a program production company. By the way, the program information of the pre-broadcast program is subjected to concealment processing such as encryption processing, and is linked to the ad space information of the ad space corresponding to the broadcast time zone of the pre-broadcast program and stored in the storage device 12. May be good.

Returning to the description of the configuration of the information processing device 10, the storage device 12 is communicably connected to the server computer 11. When providing this service, the server computer 11 accesses the storage device 12, reads out various information stored in the storage device 12, and performs a series of processes using the information.

In the configuration shown in FIG. 3, the storage device 12 is separated from the server computer 11, and is configured by, for example, a computer different from the server computer 11, specifically, a database server. However, the storage device 12 is not limited to this, and the storage device 12 may be configured by an auxiliary storage device (that is, a storage 11d) installed in the server computer 11 or externally attached to the server computer 11.

Further, as shown in FIG. 3, the server computer 11 is communicably connected to the terminal (client terminal 14) of the service user T via the network 13. As described above, the client terminal 14 inputs information necessary for using this service, specifically, information related to the broadcast time zone of the advertisement to be produced in the future. Information related to the broadcast time zone of the advertisement to be broadcast in the future includes, for example, the program name of the program to be broadcast in the broadcast time zone (the program to be broadcast in the future) and other meta information in the advertisement to be broadcast in the future. Is done. The input information for using this service will be described in detail later.

The input information is transmitted from the client terminal 14 to the server computer 11, and the server computer 11 generates the image information of the above advertisement based on the input information received from the client terminal 14. Finally, the server computer 11 outputs the generated image information, and the output information is displayed on the screen of the client terminal 14.

<< About the function of the information processing device of this embodiment >>
Next, the function of the information processing device 10 will be described with reference to FIG. FIG. 4 is a block diagram showing a functional unit included in the information processing device 10.

As shown in FIG. 4, the information processing device 10 includes an input unit 21, a storage unit 22, an extraction unit 23, a learning unit 24, a generation unit 25, a verification unit 26, a calculation unit 27, and an output unit 28. Of these functional units, the storage unit 22 is realized by the storage device 12. Functional units other than the storage unit 22 are installed in the hardware devices (that is, the processor 11a, the memory 11b, the communication interface 11c, the storage 11d, the input device 11e and the output device 11f) included in the server computer 11 and the server computer 11. It is realized by collaborating with the service program.

The input unit 21 acquires the input information input by the service user T on the client terminal 14 when using this service. The input information includes information related to the broadcast time zone of the advertisement to be produced in the future, for example, the inventory information corresponding to the broadcast time zone of the advertisement to be produced in the future, or the program to be broadcast in the broadcast time zone. Includes program information.

In the present embodiment, the input unit 21 acquires the following information i1 to i5 as input information.
i1: Program information about programs that will be broadcast during the broadcast time of future advertisements i2: Product information about the types of products or services that will be advertised in future advertisements i3: About the content of future advertisements Keyword i4: Meta information of pre-broadcast programs to be broadcast in the future i5: Number of outputs for image information of advertisements to be produced in the future

Specific examples of the information i1 include a program genre, a broadcast time zone, a broadcast day of the week, a program name, a context caused by viewing, and the like.
Specific examples of the information i2 include categories such as products advertised in advertisements.
Specific examples of the information i3 include, for example, the name of the talent used in the advertisement, a word expressing the general concept of the advertisement, and the like.
Specific examples of the information i4 include, for example, the broadcast time zone or the program name of the program.

The input information is not limited to the above information i1 to i5, and may include information other than these information, and also includes a part of the above information i1 to i5 (for example, information i5). It does not have to be. For example, the number of programs for which the degree of matching with advertisements to be broadcast in the future is calculated by the calculation unit 27, and the set value (required value) of the viewing amount such as GRP (Gross Rating Point) or the number of reach are included. You may.

The storage unit 22 is composed of a storage device 12, and stores various types of information stored in the storage device 12 in a database. Specifically, as shown in FIG. 4, the storage unit 22 includes a first database 31 containing acquired program information and acquired advertisement information, a second database 32 containing ad space information, and broadcasting. A third database 33 containing the program information of the previous program is configured.

In the present embodiment, the program information of each program recorded in the third database 33 is associated with the information of any one of the advertising space information recorded in the second database 32. .. That is, when one of the ad space information recorded in the second database 32 is specified, the pre-broadcast program corresponding to the ad space information is determined, and the information of the pre-broadcast program is selected from the third database 33. Become. On the contrary, when one of the program information recorded in the third database 33 is specified, the advertising space corresponding to the program information is determined, and the information of the advertising space is selected from the second database 32.

The extraction unit 23 identifies the specified program based on the input information acquired by the input unit 21. The specific program is, for example, a program specified from the above information i1, and specifically, a program corresponding to the program name or context indicated by the information i1. The specific program may be a program broadcast in the past or a program broadcast in the future.

Further, the extraction unit 23 obtains the acquired advertisement information (strictly speaking, the meta information of the advertisement) of the advertisement matching with the specific program based on the acquired program information and the acquired advertisement information recorded in the first database 31. Extract. Explaining with a specific example, the extraction unit 23 selects an advertisement according to the input information acquired by the input unit 21 from among a plurality of advertisements for which the acquired advertisement information has been acquired. For example, the above information i2 Select advertisements that match the categories of the products shown (hereinafter referred to as category-matching advertisements). After that, the extraction unit 23 digitizes (vectorizes) the advertisement information of the selected category-matching advertisement and the acquired program information of the specified program, and obtains the degree of matching between both numerical values (vectors). Finally, the extraction unit 23 extracts the advertisement information of n category-matched advertisements (n is a natural number) in order from the one with the highest degree of matching.

The number n of the advertisement information of the category-matched advertisements to be extracted is not particularly limited, but is preferably a sufficient number for carrying out machine learning described later.

The learning unit 24 learns the tendency of the acquired advertisement information (that is, the advertisement information of the category matching advertisement) extracted by the extraction unit 23. More specifically, the learning unit 24 digitizes (vectorizes) the advertisement content and performers indicated by the extracted acquired advertisement information, and performs machine learning to identify the tendency of the value (vector). To carry out. Here, learning about trends refers to words that frequently appear among acquired advertisement information that has a high degree of matching with a specific program, words that are conceptually similar, and words that have a high degree of co-occurrence with these words. It is to learn the tendency of, and specifically, to construct a mathematical model (hereinafter, an estimation model) that estimates the above words from the information of the program.

Machine learning includes known learning algorithms such as ID3 (Itterative Dichotomiser 3), correlation rule learning, neural networks, convolutional neural networks, recurrent neural networks, deep learning neural networks, attentions, transformers, hostile generation networks, and inheritance. Algorithmic programming, functional logic programming, support vector machine, clustering, principal component analysis, cluster analysis, Boltzmann machine, Basin network, decision tree, Rankanashi forest, and extreme learning machine are available and can be developed in the future. Learning algorithms are also available. Further, as a method for quantifying (vectorizing) the advertisement content and the like indicated by the advertisement information, known methods such as word2vec and LDA (Latent Dirichlet Allocation) can be used.

The generation unit 25 generates (appropriate) image elements according to the broadcast time zone of the advertisement to be produced in the future, based on the input information acquired by the input unit 21. An image element is a component of an image of an advertisement to be produced in the future. Specifically, a language (word or sentence), an image (still image or video), a voice (melody or voice), etc. representing the image are included. Applicable. In the present embodiment, it is assumed that the image element is a language, particularly a word group consisting of a plurality of words.

In the present embodiment, as shown in FIG. 4, the generation unit 25 is composed of a first generation unit 25a and a second generation unit 25b. The first generation unit 25a generates a word group corresponding to the input information as an image element by using the tendency learned by the learning unit 24. More specifically, the first generation unit 25a generates a word group according to the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future, based on the tendency learned by the learning unit 24.

To give an example of a specific procedure, a program based on input information (a program broadcast in the past or a program broadcast in the future) is identified, and words matching the identified program are learned based on the tendency to be learned. To find out. Determining based on the learned tendency is, for example, inputting information on a program to be broadcast in the broadcast time zone of an advertisement to be produced in the future into the above-mentioned estimation model, and a characteristic word that matches the program ( Specifically, performers, scenes, scenes, etc.) are derived as image elements.

Further, in the present embodiment, a keyword related to the content of the advertisement to be produced in the future is specified from the information i3 in the input information, and the specified keyword is adopted as an image element. That is, the first generation unit 25a generates a word group including the above keywords as an image element. The number of words in the word group is not particularly limited, but may be, for example, 100 or more words.

The second generation unit 25b searches for a natural word combination including a keyword related to the content of the advertisement to be produced in the future from the word group generated by the first generation unit 25a. A natural word combination is a combination of words that can be established (in other words, a combination other than a combination of words that cannot co-occur in a sentence), for example, a combination of words with a high degree of co-occurrence. , It can be searched by applying a known natural language processing technique (specifically, co-occurrence degree analysis). In the following, natural word combinations will be referred to as "word sets" for convenience.

In the present embodiment, the second generation unit 25b searches for a plurality of word sets to generate an image element. At this time, the same word (strictly speaking, a word other than the keyword) may be duplicated in the word set.

The verification unit 26 executes a verification process for a word group (specifically, a plurality of word sets generated by the second generation unit 25b) as an image element generated by the generation unit 25. The verification process is a process of verifying the degree of matching between the image element generated for the advertisement to be produced in the future and the program broadcast in the broadcast time zone of the advertisement. In the verification process, the verification unit 26 refers to the program information of the above program in the third database 33, and digitizes (vectorizes) the program information and the image element to obtain the matching degree. As a method for calculating the degree of matching and a method for digitizing (vectorizing) image elements and program information, the above-mentioned known methods can be used.

Then, when the matching degree verified by the above verification process satisfies the setting condition, the image element (word set) is adopted as image information and output by the output unit 28. The setting condition is, for example, a condition that the matching degree is equal to or higher than the reference value.

In this embodiment, a number of image information corresponding to the number of outputs indicated by the information i5 included in the input information is to be output, and as a premise, the same number of image elements (word sets) are required. Become. Therefore, it is determined whether the number of word sets whose matching degree verified by the verification unit 26 satisfies the setting condition is the same as the number of outputs described above, and when the number of word sets is less than the number of outputs, the generation unit 25 The image element will be generated again by.

The calculation unit 27 executes a calculation process for each of a plurality of image elements (word sets) whose matching degree verified by the verification unit 26 satisfies the setting condition. The calculation process is a process of calculating the degree of matching between each of a plurality of pre-broadcast programs to be broadcast in the future and an image element (word set). In the calculation process, the calculation unit 27 selects a plurality of pre-broadcast programs to be broadcast in a predetermined period in the future, refers to the program information of each of the plurality of pre-broadcast programs in the third database 33, and refers to the program information. The degree of matching is obtained by digitizing (vectorizing) each of the image elements. As an example of a plurality of pre-broadcast programs to be broadcast in a predetermined period in the future, as described above, all the programs broadcast on the television broadcasting station J on each day of the week including the broadcast date of the advertisement to be produced in the future. The program is mentioned.

The output unit 28 outputs image information representing an image of an advertisement to be produced in the future by the image element generated by the generation unit 25. In the present embodiment, the output unit 28 outputs image information including the word set, which is an image element, when the matching degree verified by the verification unit 26 satisfies the setting condition.

Further, as shown in FIG. 5, the output unit 28 outputs a number of image information corresponding to the number of outputs indicated by the information i5 included in the input information. FIG. 5 is a diagram showing an output example of image information, and FIG. 5 shows a screen on which five image information are displayed.

Further, in the present embodiment, for each of the output image information, the calculation unit 27 calculates the degree of matching between the image element (word set) included in the image information and each of the plurality of pre-broadcast programs, and the output unit 27 calculates. 28 outputs the calculated matching degree for each image information, as shown in FIG.

The output method of the image information or the like by the output unit 28 is not particularly limited, and for example, data for displaying the image information or the like on the screen is transmitted to the client terminal 14. , It is conceivable to execute a process for printing image information or the like with a printer or the like.

<< About the information processing flow of this embodiment >>
Next, as the information processing flow of the present embodiment, the processing flow for providing the service will be described with reference to FIGS. 6 and 7. 6 and 7 are diagrams showing the flow of the processing flow for providing the service.
In the following description, the operation by the server computer 10 constituting the information processing device 10 will be mainly described. That is, in the following, each step in the processing flow will be described as an operation example of the server computer 11.

In addition, the following description includes a description of the information processing method of the present invention. That is, the information processing method of the present invention is adopted in the processing flow for providing the service, and each step in the processing flow corresponds to a step constituting the information processing method of the present invention.

In the following description, it is assumed that the advertisement to be produced in the future is the advertisement X and the advertisement space which is the broadcasting time zone of the advertisement X is the advertisement space Y.

In the processing flow for providing this service, first, the server computer 11 communicates with the client terminal 14 and acquires the information related to the advertisement X and the advertisement space Y, specifically, the input information including the information i1 to i5. (S001). At the time of this step S001, the advertiser or the like who is the service user T operates the client terminal 14 to input the contents of the information i1 to i5, and after the input is completed, transmits the input information to the server computer 11.

Next, the server computer 11 selects the specified program according to the input information acquired in step S001 from among the plurality of programs in which the program information (strictly speaking, the acquired program information) is stored in the first database 31. Identify (S002). Specifically, the program corresponding to the program name or context indicated by the information i1 in the input information is defined as the specified program.

Next, the server computer 11 extracts the acquired advertisement information of the advertisement matching with the specific program based on the information recorded in the first database 31 (S003). Specifically, an advertisement (category-matched advertisement) that matches the category of the product or the like indicated by the information i2 in the input information is selected, and among the selected category-matched advertisements, the degree of matching with the specified program is high n. The advertisements are found, and the information of the n advertisements (strictly speaking, the meta information included in the acquired advertisement information) is extracted from the first database 31.

Next, the server computer 11 learns the tendency of the meta information of the category matching advertisement extracted in step S003 by machine learning (S004). In this step S004, an estimation model for estimating a word matching with the specified program is constructed from the information of the specified program.

Next, the server computer 11 generates an image element according to the advertising space Y based on the input information based on the tendency learned in step S004 (that is, the estimation model) (S005). In step S005, the server computer 11 generates a word group including the keyword of the advertisement X indicated by the information i3 in the input information. Specifically, the program corresponding to the advertising space Y (hereinafter, program Z) is specified from the input information, the information of the program Z is input to the estimation model, and the language matching with the program Z is derived as an image element.

Next, the server computer 11 searches for and generates a natural word combination (word set) including the keyword of the advertisement X indicated by the information i3 from the image element generated in step S005, that is, the word group (S006). ). In this step S006, the server computer 11 searches for and generates a plurality of word sets.

Next, the server computer 11 verifies the degree of matching between the word set and the program Z for each of the plurality of word sets generated in S006 (S007). Then, the server computer 11 determines whether or not there is a word set in which the verified matching degree satisfies the setting condition (that is, a word set in which the matching degree is equal to or higher than the reference value) (S008). Further, the server computer 11 determines whether or not the number of word sets whose verified matching degree satisfies the setting condition is equal to or greater than the number of outputs indicated by the information i5 in the input information (S009).

If both step S008 and step S009 are Yes, the process proceeds to the next step S010. On the other hand, if any one of step S008 and step S009 is No, the process returns to step S005, and the server computer 11 regenerates the image element.

In step S010, the server computer 11 calculates the matching degree between the word set and the pre-broadcast program to be broadcast in the future for each of the plurality of word sets whose verified matching degree satisfies the setting condition. Specifically, a plurality of pre-broadcast programs to be broadcast in a predetermined period in the future are selected, and the degree of matching between each pre-broadcast program and each word set is obtained using the program information of each selected pre-broadcast program. ..

After that, the server computer 11 outputs the image information of the advertisement X including the word set in which the verified matching degree satisfies the setting condition (S011). In this step S011, the server computer 11 outputs the same number of image information as the number of outputs indicated by the information i5 (see FIG. 5).
Further, as shown in FIG. 2, the server computer 11 outputs the degree of matching between the word set calculated in step 010 and each pre-broadcast program for each word set (that is, image information) (S012).

When the series of steps described above are completed, the processing flow for providing this service ends.
After that, when the service user T inputs new input information, the above processing flow is executed again with the new input information as a trigger, and the above series of steps are repeated.

<< About the effectiveness of this embodiment >>
In the present embodiment, as described above, input information including information related to the broadcast time zone of the advertisement to be produced in the future is acquired, and based on the input information, it corresponds to the broadcast time zone of the advertisement to be produced in the future. An image element is generated, and the generated image element outputs image information representing an image of an advertisement to be produced in the future.

With the above configuration, in the present embodiment, it is possible to propose an image suitable (suitable) for the broadcasting time zone of the advertisement to be produced in the future and support the advertisement production. That is, in the present embodiment, the image of the advertisement matching (high affinity) with the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future is output, so that the service user T sees the image and sees the image. It can be used as a reference for advertising production.

Further, in one embodiment of the present invention, the input unit uses program information related to a program to be broadcast during the broadcast time zone of an advertisement produced in the future and a product related to a type of product or service advertised in the advertisement produced in the future. It is advisable to acquire input information including such information, keywords related to the contents of advertisements to be produced in the future, and meta information of programs to be broadcast in the future.
If the above-mentioned information is acquired as input information, more accurate image elements can be generated based on the above-mentioned input information for the image of the advertisement to be produced in the future.

Further, in the present embodiment, the input information includes the evaluation (context) information i1 regarding the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future. Thereby, for example, it is possible to propose an image of an advertisement that matches the context indicated by the information i1.

Further, in the present embodiment, the degree of matching between the generated image element and the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future is verified, and when the verified degree of matching satisfies the setting condition, the said Outputs image information including image elements. As a result, the image information of the advertisement to be produced in the future can be proposed after confirming its validity, so that the credibility of the proposed image information is improved.

Further, in the present embodiment, for the image element whose matching degree satisfies the setting condition, the matching degree with each of the plurality of pre-broadcast programs to be broadcast in the future is calculated, and the calculation result is output. As a result, it is possible to find a time zone suitable for broadcasting the advertisement to be produced in the future from each broadcast time zone of the plurality of pre-broadcast programs. As a result, the service user T can find an appropriate ad space for the advertisement of the proposed image.

<< Other Embodiments >>
Although the information processing apparatus and the information processing method of the present invention have been described above with specific examples, the above-described embodiment is merely an example, and other embodiments can be considered. For example, in the above-described embodiment, it is decided to acquire program information related to a program to be broadcast in the broadcast time zone as information related to the broadcast time zone of the advertisement to be produced in the future. Then, in the above-described embodiment, it is decided to generate an image element (word group) corresponding to the above-mentioned program. However, the present invention is not limited to this, and information on the ad space corresponding to the broadcast time zone may be acquired as information related to the broadcast time zone of the advertisement to be produced in the future. Image elements may be generated according to the inventory.

Further, in the above-described embodiment, the degree of matching between the generated image element (word set) and the pre-broadcast program to be broadcast in the future is calculated, and the calculation result is output. However, the present invention is not limited to this, and the step of calculating the degree of matching between the generated image element and the pre-broadcast program may be omitted.

Further, in the above-described embodiment, it was decided to verify the degree of matching between the generated image element (word set) and the program to be broadcast in the broadcast time zone of the advertisement to be produced in the future. Then, in the above-described embodiment, when the verified matching degree is equal to or higher than the reference value, the image information including the image element is output. However, the present invention is not limited to this, and the step of verifying the degree of matching between the generated image element and the program broadcast in the broadcast time zone of the advertisement produced in the future may be omitted.

Further, in the above-described embodiment, a plurality of words (word groups) including keywords of advertisements to be produced in the future are generated as image elements according to the input information, but the present invention is not limited to this. For example, only one word may be generated as an image element.

10 Information processing device 11 Server computer 11a Processor 11b Memory 11c Communication interface 11d Storage 11e Input device 11f Output device 12 Storage device 13 Network 14 Client terminal 21 Input unit 22 Storage unit 23 Extraction unit 24 Learning unit 25 Generation unit 25a First generation Part 25b 2nd generation part 26 Verification part 27 Calculation part 28 Output part 31 1st database 32 2nd database 33 3rd database 100 Service provision system J TV broadcasting station K Research company S Service provider T Service user

Claims (7)

An input unit that acquires input information including program information related to programs that will be broadcast during the broadcast time of advertisements that will be produced in the future.
A generator that generates an image element according to the broadcast time zone based on the input information,
An output unit that outputs image information representing the image of the advertisement to be produced in the future by the image element, and
A storage unit that stores acquired program information acquired for each of a plurality of programs and acquired advertisement information acquired for each of a plurality of advertisements broadcast in the past.
The specified program is specified based on the input information, and the acquired advertisement information of the advertisement matching with the specified program is extracted based on the acquired program information and the acquired advertisement information stored in the storage unit. Extractor and
It has a learning unit that learns the tendency of words that frequently appear among the extracted acquired advertisement information, words that are conceptually similar, and words that have a high degree of co-occurrence with these words.
The information processing device generates, as the image element, a characteristic word that matches a program to be broadcast in the broadcast time zone of the advertisement to be produced in the future from the learned tendency and the input information. .. The input unit is
Program information about programs that will be broadcast during the broadcast hours of the advertisements that will be produced in the future,
Information on products, etc. related to the types of products or services advertised in the advertisements produced in the future,
Keywords related to the content of the advertisement to be produced in the future,
Meta information of programs to be broadcast in the future and
The information processing apparatus according to claim 1, wherein the input information including the above is acquired. The information processing device according to claim 2, wherein the program information includes evaluation information about the program. The information processing device according to claim 2 or 3, wherein the generation unit generates a plurality of words including the keyword as the image element. Has a the image element that has been made raw, the verification unit to verify the degree of matching of the program to be broadcast during the broadcast time period of the advertisement to be produced the future,
The information processing apparatus according to any one of claims 1 to 4 , wherein when the verified degree of matching satisfies the setting condition, the output unit outputs the image information including the generated image element. .. The information processing according to claim 5 , further comprising a calculation unit for calculating the matching degree between the image element whose matching degree verified by the verification unit satisfies the setting condition and each of a plurality of programs to be broadcast in the future. apparatus. The computer
Acquires input information including program information related to programs that will be broadcast during the broadcast time of future advertisements.
Based on the input information, an image element corresponding to the broadcast time zone is generated.
Wherein the image element, and outputs the image information representing an image of the advertisement to be produced the future,
The acquired program information acquired for each of the plurality of programs and the acquired advertisement information acquired for each of the plurality of advertisements broadcasted in the past are stored in the storage unit.
The specific program is specified based on the input information, and the acquired advertisement information of the advertisement matching with the specific program is extracted based on the acquired program information and the acquired advertisement information stored in the storage unit. ,
Learn the tendency of words that frequently appear among the extracted acquired advertisement information, words that are conceptually similar, and words that have a high degree of co-occurrence with these words.
An information processing method that generates, as the image element, a characteristic word that matches a program broadcast in the broadcast time zone of the advertisement to be produced in the future from the learned tendency and the input information .