JP2020086405A - Information processor, information processing system and information processing program - Google Patents

Abstract

To shorten time for replacing a character matching a specific replacement condition with a character having a different character shape.SOLUTION: An information processor 1 includes: an image data generation section 2 for generating image data of a second character on the basis of a character image indicating an image of the second character having a character shape different from a character shape of a first character described in a sentence; and a character replacement section 3 for replacing the first character with the second character on the basis of the image data of the second character when a character positioned before or after the first character matches a specific replacement condition in the sentence.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information processing device, an information processing system, and an information processing program.

Patent Document 1 discloses a technique that enables printing in a variant character shape desired by a user even in the case of text printing by an application that does not have a function of switching the character shape. In the technique disclosed in Patent Document 1, a character code having a plurality of variant data is detected from print data, the plurality of variant characters are displayed as candidate glyphs on the screen of the display unit, and displayed on the screen of the display unit. When the user refers to the candidate glyph and selects the glyph of the character to be used for printing, replacement with a character having a different glyph is performed.

However, with the technique typified by Patent Document 1, it is not possible to collectively change only the characters that meet a specific replacement condition into characters with different glyphs. Therefore, the user needs to select only the characters that meet a specific replacement condition, and thus the larger the number of characters to be replaced, the greater the burden on the user and the long working time. ..

In view of the above problems, the present invention shortens the time for replacing a character that matches a specific replacement condition with a character having a different glyph.

In view of the above problem, the information processing apparatus according to the present invention generates image data of the second character based on a character image showing an image of a second character having a character shape different from the character shape of the first character described in the sentence. In the image data generation unit and in the sentence, when a character positioned before or after the first character matches a specific replacement condition, the first character is changed to the second character based on the image data of the second character. And a character replacement unit for replacement.

According to the present invention, it is possible to reduce the time required to replace a character that matches a specific replacement condition with a character having a different glyph.

It is an example of a functional block diagram of the information processing apparatus according to an embodiment of the present invention. It is a figure which shows the difference of the character shape of a font. 7 is a flowchart for explaining the operation of the information processing device according to the present embodiment. It is a figure which shows the image before and after the replacement of the character image by an image variable program. It is FIG. 1 which shows the image replaced by the image of image variable data. It is FIG. 2 which shows the image replaced by the image of image variable data. It is a figure for explaining a mode that a part of character described in a sentence is replaced. It is an example of a functional block diagram of the information processing apparatus 1 according to the first modification. 9 is a flowchart for explaining the operation of the information processing device 1 according to the first modification. It is an example of a functional block diagram of the information processing apparatus 1 according to the second modification. 11 is a flowchart for explaining the operation of the information processing device 1 according to the second modification. It is an example of a functional block diagram of the information processing apparatus 1 according to the third modification. It is a flow chart for explaining operation of information processor 1 concerning the 3rd modification. It is a figure which shows the hardware structural example for implement|achieving the information processing apparatus 1 which concerns on embodiment of this invention. It is a figure which shows the structural example of 1 A of information processing systems provided with the information processing apparatus 1 which concerns on this Embodiment. It is a figure which shows an example of the machine learning part 400 applied to the replacement condition proposal part 4 or the replacement condition proposal part 41.

Embodiments for carrying out the present invention will be described below with reference to the drawings.

Embodiment.
FIG. 1 is a diagram showing an example of functional blocks of an information processing apparatus according to an embodiment of the present invention. The information processing device 1 is a PC (Personal Computer) or the like. The information processing device 1 includes an image data generation unit 2, a character replacement unit 3, a first storage unit 10, and a second storage unit 20. The image data generation unit 2 generates the image data of the second character by the user based on the character image that is generated by the user and shows the image of the second character having a character shape different from the character shape of the first character described in the sentence. The character replacement unit 3 determines whether the character positioned before or after the first character in the sentence matches a specific replacement condition, and as a result of the determination, the character positioned before or after the first character is When the specific replacement condition is met, the first character is replaced with the second character based on the image data of the second character. The specific replacement condition will be described later. The text data in which the characters have been replaced by the character replacement unit 3 is stored in a memory (not shown).

For example, when a plurality of glyphs conforming to “JIS 90” are described in a document created in the past, the information processing apparatus 1 selects only those glyphs that meet a specific replacement condition from “glyph”. It is configured so that it can be changed to a character shape according to JIS 2004. It should be noted that the types of glyphs before and after the change are not limited to these. Below, the conventional problems and the shape of the font will be described, and then the operation of the information processing apparatus 1 will be described.

There is a demand to collectively change the glyphs of characters included in documents created in the past. The glyph means the appearance (design) of a character having a specific skeleton. Even if the characters have the same design, the number of lines that form a character, the thickness of this line, the number of points that form the character, the angle of this point, etc. may differ, so the line that forms the character, the point, etc. It can be said that different characters have different glyphs. For example, representative standards for character sets include "JIS X 0208" revised in 1990 and "JIS X 0213" revised in 2004. "JIS X 0208" is a Japanese industrial standard that defines a 2-byte coded character set for information exchange, including graphic characters used in Japanese notation, place names, personal names, and the like. The standard name of "JIS X 0208" is "7-bit and 8-bit 2-byte information exchange coded kanji character set". "JIS X 0208" corresponds to "JIS 90" described above. "JIS X 0213" is a Japanese Industrial Standard that extends "JIS 90" and defines a coded character set for Japanese. The standard name of “JIS X 0213” is “7-bit and 8-bit 2-byte information exchange encoded extended kanji character set”. "JIS X 0213" corresponds to "JIS 2004" described above.

Both "JIS 2004" and "JIS 90" are standards for character sets, not standards for defining glyphs. The glyphs used in the standards are merely examples, not norms. However, each company tends to respect the exemplary glyphs in the standard when creating fonts, and as a result, there are fonts according to “JIS 2004” and fonts according to “JIS 90”. An example of these fonts is shown in FIG.

FIG. 2 is a diagram showing a difference in font shape. On the left side of FIG. 2, the font shape according to “JIS 2004” is shown. On the right side of FIG. 2, the font shape according to "JIS 90" is shown. When two types of kanji representing "tsuji" are compared, the font shape according to "JIS 2004" is different from the font shape according to "JIS 90". Regarding the two types of kanji representing "Katsu", the font shape according to "JIS 2004" is different from the font shape according to "JIS 90".

Here, for example, if there are multiple glyphs conforming to "JIS 90" in a document created in the past, you want to change all of these glyphs to conform to "JIS 2004", that is, replace them. There is a demand that. If both "JIS 90 font" and "JIS 2004 font" have the same glyph for all characters except the character to be replaced, the font name specified in the document is changed from "JIS 90 font" to "JIS You can change all the glyphs in the text at once by simply changing to "2004 font". However, among the "JIS 90 fonts" and "JIS 2004 fonts", those provided with the same design are limited to some fonts such as "MS Mincho" and "MS Gothic". Therefore, in many cases, the "JIS 2004 character" font of the same design is not provided to the "JIS 90 character" font used in the past. In this case, in order to collectively change the "JIS 90 glyph" to the "JIS 2004 glyph", it was necessary to prepare the "JIS 2004 glyph" font independently. However, it takes a lot of money and time to create a font, so it is not easy to change all the glyphs.

In addition, when the font of "JIS 90 glyph" is used in the document created in the past, only the glyph of the character that meets the specific replacement condition among the plurality of characters in this document is specified in "JIS 2004 There is also a desire to change to "glyphs". For example, it is assumed that the names "Tsuji Ataro" and "Tsuji Boko" are mixed in the same document. In addition, this document uses a "JIS 90 glyph" font, and it is assumed that all kanji that represent "tsuji" are single-point Shinji "tsuji." Here, "Tsuji" of "Tsuji Boko" is changed to "Tsuji" of two points based on his strong hope, and "Tsuji" of "Taro Tsuji" is one point based on his wish. The case of leaving it as it is will be described.

This document shall have the following features:

The first feature is that "Tsuji Ataro" may be written as "Tsuji visiting professor" in the same document. It is also sometimes referred to as "visiting professor Tsuji". In this case, "Visiting Professor" must exist within 10 characters before or after "Tsuji".

As a second feature, "Boji Tsuji" may be described as "Professor Emeritus Tsuji" in the same document. It is also sometimes referred to as "Professor Emeritus Tsuji". In this case, "Professor Emeritus" must exist within 10 characters before and after "Tsuji".

As a third feature, the description about “Taro Tsuji” and the description about “Boko Tsuji” are described separately from each other in the text, for example, within 10 characters before and after “Tsuji” "And "Professor Emeritus" shall never exist. Furthermore, it is assumed that there is no "professor emeritus" within 10 characters before and after "Taro Tsuji" and there is no "visiting professor" within 10 characters before and after "Boko Tsuji".

In such a characteristic sentence, only when a specific replacement condition is satisfied, the single point Shinji "Tsuji" of the character string "Tsuji Boko" is changed to the two-point Shinji "Tsuji". Need to respond to requests. For example, if "Professor Emeritus" exists within 10 characters before or after the character string "Tsuji Biko", or within 10 characters before or after "Tsuji", then one point is two Shinji "Tsuji". The condition is that it is changed to Shinji's "Tsuji", and in all other cases, it remains as a single point Shinji's "Tsuji".

There are similar requests for proper nouns other than personal names. For example, there may be variants in the characters used in the title of a work, the title of a song, and the like. For example, there is a document explaining the Daibosatsutoge Pass, which is a place name in Yamanashi Prefecture. This document assumes that "JIS 90 glyph" fonts are used. There is a part of this document that refers to Nakazato Uesan's novel "The Great Bodhisattva Pass". However, due to the strong demands of the author's bereaved family, it is necessary to change only the "satsu" in the novel name "Daisatsusatsutoge" to "JIS 2004" "satsu". And, in this document, the novel name "Daibosatsu Pass" is always written in the form of "Novel "Daisatsusatsu Pass", "Daisatsu Daisatsusugi Novel", and the character string "Daisatsusatsu Pass" It is assumed that the character string "novel" always exists within 7 characters before and after. That is, when the character string "novel" exists within 7 characters before and after the character string "daibosatsutoge", "daibosatsutoge" always represents the novel name. In this case, only when the character string "novel" exists within 7 characters before and after "Daibatsusatsutoge", the "satsu" that exists within 7 characters before and after the character string "novel" is converted into "JIS 2004 character form". It is necessary to change to "Satsu" and to respond to the request that "Satsu", which does not exist within 7 characters before and after the character string "novel", be left as "JIS 90 character form".

In this way, when changing only the glyphs of the characters that match the specific replacement condition, the conventional technique typified by Patent Document 1 does not include a unit for determining the specific replacement condition. It was difficult to change only the glyphs of the matching characters all at once. In order to solve such a problem, the information processing apparatus 1 according to the present embodiment determines a character shape of a specific character in a sentence when a character positioned before or after the character matches a specific replacement condition. It is configured to replace with a different letter. The operation of the information processing device 1 will be described below.

FIG. 3 is a flow chart for explaining the operation of the information processing apparatus according to this embodiment. In step S1, the user generates character image data for an image variable program. The image variable program is a program for executing variable printing. Variable printing is to change the content to be printed based on input data and perform printing. Variable printing can be considered to be a high-performance version of "mail merge printing" performed in the office. The image variable program is a program that embeds characters as images in photographs, paintings, etc. by using a variable printing method. It is mainly used for calendars with names and direct mail. In the present embodiment, the image variable program functions as the image data generation unit 2 shown in FIG.

In this embodiment, for example, "FusionPro Expression" (registered trademark) is used for the image variable program. “Fusion Pro Expression” (registered trademark) is variable printing software that combines data and images to generate personalized images. According to “Fusion Pro Expression” (registered trademark), it is possible to insert a unique image prepared by the user and create a unique original design. The image variable program may be any program as long as it is a program for executing variable printing, and is not limited to “FusionPro Expression” (registered trademark).

Some image variable programs have an "Image Character" function. This is to replace the glyphs of each character forming the character string with an image prepared for each character and output the image. FIG. 4 is a diagram showing an image before and after replacement of a character image by the image variable program. FIG. 4 shows an example in which the characters “M”, “A”, and “X” in the text are replaced with the corresponding character images. The character on the left side of FIG. 4 is the character in the sentence, and the character on the right side of FIG. 4 is the character replaced by the character image. For the character image, an image format file such as PNG (one of file formats for compressing and recording image data) and TIF is prepared for each character. By giving a file name of "character code.extension" to the image format file of characters such as "M", "A", and "X", the image variable program can identify the character image file. For example, the character image file of "M" is created by giving a file name such as "M character code.png" or "M character code.tif" to the image format file of the character "M". To be done. The data created in this way is the character image data for the image variable program described above.

The character image file handled by the image variable program is often a decorative character, but the character image file used by the information processing device 1 according to the present embodiment is not a decorative character but a normal font.

Returning to FIG. 3, in step S2, the image data generation unit 2 generates image variable data using the character image data for the image variable program.

The above-mentioned "FusionPro Expression" (registered trademark) is combined with "FusionPro VDP CREATOR" (registered trademark). “FusionPro VDP CREATOR” (registered trademark) is an example of a variable print program (an example of a merge print program), and functions as the character replacement unit 3 shown in FIG. In this embodiment, the variable print program functions as the character replacement unit 3 shown in FIG.

In step S3, the character replacement unit 3 determines, with respect to a specific character in the sentence, whether or not the character positioned before or after the character matches the specific replacement condition. When the character positioned before or after the character matches the specific replacement condition (step S3, Yes), the process of step S4 is executed. When the character positioned before or after the character does not meet the specific replacement condition (No in step S3), the processing operation ends without the character being replaced.

The details of the method of determining whether or not the character positioned before or after the specific character matches the specific replacement condition in step S3 will be described later.

In step S4, the character replacement unit 3 replaces all the characters in the designated field (text frame) in the document with the image of the image variable data. That is, the character replacement unit 3 is configured to replace only specific characters in a sentence with an image of image variable data. An image replacement method will be described with reference to FIGS.

FIG. 5 is a first diagram showing an image to be replaced with an image of image variable data. The text frame TF in the document is specified on the screen 100A on the left side of FIG. The background image BGI is specified on the screen 100B on the right side of FIG. After the text frame TF and the background image BGI are set in this way, the variable print program is executed. Then, the variable print program replaces the "Thessa" portion of the text frame TF with the character string (such as "MAX") read from the merge data (such as the list of names), and further, the characters "M" that make up the character string , "A", "X" are replaced with the corresponding images. In this way, the variable printing program searches the document for the character corresponding to the character image file created by the user, and replaces all the found characters with the corresponding image. In addition, for example, the JIS code is used for the character search, but the search is not limited to the JIS code. For example, the JIS code of the character in the sentence is compared with the JIS code of the character changed to the image variable data, and the image of the character having the matching JIS code is replaced.

FIG. 6 shows an example of the character image replaced by executing the variable printing program. FIG. 6 is a second diagram showing an image to be replaced with an image of image variable data. As shown in FIG. 6, the text frame TF designated on the screen 100A of FIG. 5 is modified. The image thus replaced is saved as a PDF file.

In the present embodiment, the image data generation unit 2 that functions as an image variable program and the character replacement unit 3 that functions as a variable print program process the entire document without the user designating the text frame TF. It is configured to be considered a subject.

Next, how a part of the characters described in the actual sentence is replaced will be described with reference to FIG. FIG. 7 is a diagram for explaining how a part of the characters described in a sentence is replaced. In step S1, character image data is created using an image creation program (paint software program or the like). The created character image data is stored in the first storage unit 10 as image image data for an image variable program.

Here, the following file names are given to the image format file of the character image data for the image variable program. The file name of "Tsuji" is "92D2.tif". The file name of "Satsu" is "8E46.tif".

In step S2, the data stored in the first storage unit 10 is converted by the image variable program into a data format (*.fpx) that can be referenced by the variable print program. The converted data is stored in the second storage unit 20 as image variable data.

In step S3, the variable print program searches the document file, and based on the information indicating the character string replacement condition, whether or not the character positioned before or after the specific character in the sentence matches the specific replacement condition. Is determined. Hereinafter, for convenience of description, the “character string replacement condition” may be simply referred to as a “replacement condition”. The information indicating the replacement condition is stored in the first storage unit 10 or the second storage unit 20. An example of the replacement condition is as follows.

For example, the first replacement condition is that if "Professor Emeritus" exists within 10 characters before and after "Tsuji Buko" and "Tsuji", replace "Tsuji" with one point that exists within 10 characters before and after "Professor Emeritus". The image variable data is replaced with "92D2.tif".

The second replacement condition is that when a "novel" exists within 7 characters before and after "Daibatsusatsutoge", the "satsu" that exists within 7 characters before and after the character string "novel" is converted to the image variable data "8E46. It is replaced with "tif".

In step S4, the character is replaced according to the determination result of step S3. The text in which the characters have been replaced is stored in PDF format in a predetermined storage unit. The character replacement method may be such that the original character is extracted and a new character is embedded in the extracted portion, or a new character may be superimposed on the original character portion.

The replacement condition is not limited to the above example. The replacement condition is that when another specific character string DEF (which may be a single character D) is detected within n characters before and after the specific character string ABC (which may be a single character A), the character string ABC is detected. (Or the single character A) may be replaced with another character string XYZ (or the single character X). n is a natural number of 1 or more.

Further, the replacement condition may be, for example, a condition based on the presence or absence of a title, the type of title, the presence or absence of a title, and the type of title. For example, "Tsuji" of 1 point Shinnyo, which is included in the character string "Professor Emeritus Tsuji" or "Professor Emeritus of Buko Tsuji", is replaced with "Tsuji" of 2 points Shinnyo. The one-point Shinji "Tsuji" included in "Tsuji visiting professor" or "Tsuji Ataro visiting professor" may remain as "one-point Shinji".

Further, the replacement condition may be, for example, a condition based on a dictionary for each field of text. For example, the single point Shinji "Tsuji" written in cooking-related texts is replaced with the two-point Shinji "Tsuji", but the single-point Shinji "Tsuji" written in music-related texts is the single point Shinji "Tsuji". It is also possible to adopt the condition of leaving it as ". In this case, dictionary data for each field is stored in, for example, the first storage unit 10, and the character replacement unit 3 refers to the dictionary data to refer to the type of vocabulary appearing in the document and the frequency of use of the vocabulary. To judge. As a result, the character replacement unit 3 includes, for example, one point Shinji "Tsuji" included in a character string such as "Tsuji Ataro's cooking class" and "Tsuji Ataro's recipe" described in the sentence. Replaced with Shinji's "Tsuji". In addition, the character replacement unit 3 includes a single point, "Tsuji", which is included in a character string such as "Tsuji A Taro's Music School" or "Tsuji A Taro's Symphony" written in a separate sentence. Leave as "Tsuji".

The characters to be replaced may be reversed by applying such a replacement condition. For example, the two-point Shinji "Tsuji" included in a specific character string written in the text of the cooking field is replaced with the one-point Shinji "Tsuji" included in the specific character string described in the text of the music field. It is possible to replace one-point Shinji "Tsuji" with two-point Shinjyo "Tsuji".

Note that it is difficult to set the replacement condition from the beginning that can obtain the replacement result that the user expects. Therefore, in order to set such a replacement condition, it takes some trial and error to improve the character replacement accuracy. It is necessary to raise it. Below, an example of composition which optimizes a substitution condition is explained.

FIG. 8 is an example of functional blocks of the information processing apparatus 1 according to the first modification. The information processing apparatus 1 illustrated in FIG. 8 includes a replacement condition proposing unit 4 and a correction receiving unit 6 in addition to the image data generating unit 2 and the character replacing unit 3.

FIG. 9 is a flowchart for explaining the operation of the information processing device 1 according to the first modification. The processing operation of steps S1 to S4 is the same as the processing operation shown in FIG.

As a result of performing the replacement in step S3 according to the above-described first replacement condition, when the replacement as expected by the user is not performed, the replacement by manual operation by the user is not performed (character) Is accepted, that is, the modification accepting unit 6 accepts the modification of the character. (Step S5).

In this way, the information indicating the content of the sentence before the characters are corrected and the information indicating the content of the sentence after the characters are corrected are associated with the information indicating the replacement condition, and are stored in the first storage unit 10 Alternatively, it is stored in the second storage unit 20.

The replacement condition proposing unit 4 compares the sentences before and after the characters are corrected for each replacement condition based on the information stored in the first storage unit 10 or the second storage unit 20, and replaces the replacement condition with the replacement condition. The correlation with the written character is analyzed (step S6).

Specifically, it is assumed that there are two substitution conditions for substituting one-point Shinji "Tsuji" with two-point Shinji "Tsuji". For example, one substitution condition (the above-mentioned first substitution condition) is that when there is a "Professor Emeritus" within 10 characters before and after "Tsuji" of 1 point Shinnyo, "Tsuji" of 1 point Shinnyo is 2 points of Shinji "Tsuji". The replacement condition is to replace with. The other replacement condition (the second replacement condition, which is a potential rule that the user is not aware of) is, for example, when a "violin" or a "violin" exists within 50 characters before or after "Tsuji" of a single-point Shinnyo. The replacement condition is to replace "tsuji" with "tsuji". The replacement condition proposing unit 4 analyzes the correlation between these replacement conditions and the replaced character.

As a result, it is possible to determine that the replacement accuracy is higher when the replacement is performed under the second replacement condition than when the replacement is performed under the first replacement condition. That is, it is possible to determine a replacement condition (for example, the second replacement condition) with few replacement errors. As a result of this determination, the replacement condition proposing unit 4 proposes the second replacement condition instead of the first replacement condition to the user as a replacement condition with few modifications (step S7).

The replacement condition proposing unit 4 notifies the user of the replacement condition proposal by the method as described below (step S8). The replacement condition proposing unit 4 may display a message indicating the contents of the replacement condition on the display unit connected to the information processing device 1, or the replacement condition proposal unit 4 may display the replacement condition from the voice output unit connected to the information processing device 1. A voice message indicating the content may be output.

FIG. 10 is an example of functional blocks of the information processing apparatus 1 according to the second modification. The information processing device 1 illustrated in FIG. 10 includes a replacement condition proposing unit 41 in addition to the image data generating unit 2 and the character replacing unit 3. The replacement condition proposing unit 41 has a function of proposing a recommended replacement condition without the user having to devise the replacement condition in order to realize an efficient character replacement work.

FIG. 11 is a flowchart for explaining the operation of the information processing device 1 according to the second modification. The processing operation of steps S1 to S4 is the same as the processing operation shown in FIG.

For example, the character replacement (correction) is manually performed in a certain range (for example, a range of ⅕ from the beginning) before and after the character replacement, and the correction receiving unit 6 receives the correction by the user. (Step S15).

In this way, the information indicating the content of the sentence before the characters are corrected and the information indicating the content of the sentence after the characters are corrected are associated with the information indicating the range in which the characters are replaced, A plurality of data are stored in the first storage unit 10 or the second storage unit 20.

By referring to these pieces of information stored in the first storage unit 10 or the second storage unit 20, the replacement condition proposing unit 41 determines whether the text content within a certain range of the text before the text is corrected and the text The content of the revised text is compared with the text content within a certain range. It is assumed that the certain range compared here is the same part of the document before and after the correction. Then, the replacement condition proposing unit 41 analyzes the correlation between the text contents before being corrected and the corrected text contents (step S16).

As a result of the analysis, the replacement condition proposing unit 41 estimates the replacement condition for executing the character replacement in a range other than the certain range in which the user replaces the character, and then the first storage unit 10 or the second storage unit 20. To store. Further, the replacement condition proposing unit 41 notifies the user of the estimated replacement condition by the method described below (step S17).

For example, a message indicating the content of the estimated replacement condition may be displayed on the display unit connected to the information processing apparatus 1, or the content of the estimated replacement condition may be displayed from the voice output unit connected to the information processing apparatus 1. A voice message may be output. As a result, the user can realize efficient character replacement work using the replacement condition estimated by the information processing device 1 without devising the replacement condition.

FIG. 12 is an example of functional blocks of the information processing apparatus 1 according to the third modification. The information processing device 1 illustrated in FIG. 12 includes a typographical error notification unit 5 in addition to the image data generation unit 2 and the character replacement unit 3. The typographical error notifying unit 5 has a function of notifying a user of a character described in the original document, if the character described in the original document has an error as a result of collating the character described in the original document with the replacement condition. ..

FIG. 13 is a flowchart for explaining the operation of the information processing device 1 according to the third modification. The processing operation of steps S1 to S4 is the same as the processing operation shown in FIG.

The typographical error notification unit 5 exists in the original document using information indicating the content of the sentence before the character is corrected, information indicating the content of the sentence after the character is corrected, and the typographical error determination condition. Error is detected (step S25).

In the above example, the typographical error condition is based on the replacement condition of replacing "Tsuji", which contains "Tsuji" with one point, with "Tsuji", which contains "Tsuji" with two points. Created.

The typographical error determination condition here is, for example, whether or not "XX" of "Professor Tsuji XX" matches "honor". The typographical error notification unit 5 confirms the content of the sentence before the characters are corrected, and when "XX" in the character string "Tsuji XX" is a character other than "honor", the character other than "honor" is displayed. For a character string that contains, for example, "Tsuji visiting professor," it is determined that the one-point score is correct.

By performing the determination process as described above, the typographical error notification unit 5 determines that one point is correct when "Tsuji" in "Tsuji visiting professor" is indicated by two points, and for example, "Tsuji visiting professor" The user is notified of a message such as "Is Tsuji all right?" (step S26). The message may be a display message or a voice message. As a result, even if "Tsuji" of "Tsuji visiting professor" in the text before the replacement of characters is written as a two-dot line, this "Tsuji" is not a two-dot line, but a one-point line. You can know that it is correct.

The function of the typographical error notification unit 5 can be combined with the information processing device 1 according to the second modification and the third modification.

FIG. 14 is a diagram showing a hardware configuration example for realizing the information processing device 1 according to the embodiment of the present invention. The information processing device 1 can be realized by a processor 101, a memory 102 including a RAM (Random Access Memory), a ROM (Read Only Memory), and an input/output interface 103 for connecting to a network. is there. The processor 101, the memory 102, and the input/output interface 103 are connected to the bus 104, and can exchange data, control information and the like with each other via the bus 104.

When the information processing apparatus 1 is realized, a program for the information processing apparatus 1 is stored in the memory 102, and the processor 101 executes the program, so that the image data generation unit 2 and the character replacement unit of the information processing apparatus 1 are executed. 3, a replacement condition proposing unit 4, a correction receiving unit 6, a replacement condition proposing unit 41, a typographical error notifying unit 5, and the like are realized. The program for the information processing device 1 is a program that executes functions such as the image data generation unit 2, the character replacement unit 3, the replacement condition proposal unit 4, the replacement condition proposal unit 41, and the typographical error notification unit 5. The input/output interface 103 is used when, for example, the character image data is stored in the first storage unit 10.

As described above, the information processing device 1 according to the present embodiment uses the image data of the second character based on the character image showing the image of the second character having a character shape different from the character shape of the first character described in the sentence. When the character positioned before or after the first character in the sentence matches the specific replacement condition, the first character is replaced with the second character based on the image data of the second character. The character replacement unit 3 is provided. With this configuration, only the characters that match the specific replacement condition can be collectively changed to the characters having different glyphs. Therefore, the user does not need to select only the characters that match a specific replacement condition, the burden on the user is significantly reduced, and the work time for character replacement is significantly reduced.

FIG. 15 is a diagram showing a configuration example of an information processing system 1A including the information processing device 1 according to the present embodiment. The information processing system 1A includes an information processing device 1, a communication network 200, and a server 300 that is a processing device. The server 300 is communicatively connected to the information processing device 1 via the communication network 200. According to the information processing system 1A, for example, the character image data stored in the server 300 can be used to replace characters, and the versatility of the information processing apparatus 1 can be enhanced.

Further, the information processing program according to the embodiment causes a computer to generate image data of a second character based on a character image indicating an image of a second character having a character shape different from the character shape of the first character described in a sentence, In the sentence, when the character positioned before or after the first character matches the specific replacement condition, the first character is replaced with the second character based on the image data of the second character. According to the information processing program according to the embodiment, only the characters that match a specific replacement condition can be collectively changed to the characters having different glyphs. Therefore, the user does not need to select only the characters that match a specific replacement condition, the burden on the user is significantly reduced, and the work time for character replacement is significantly reduced.

FIG. 16 is a diagram showing an example of the machine learning unit 400 applied to the replacement condition proposing unit 4 or the replacement condition proposing unit 41. Machine learning is a technology for allowing a computer to acquire human-like learning ability.The computer autonomously generates algorithms necessary for judgment such as data identification from learning data that is captured in advance, and This is a technique for applying predictions to various data to make predictions. An example of the machine learning method will be described below.
(Supervised learning) A method of identifying data and deriving the rules based on sample data given in advance (eg regression analysis, SVM (Support Vector Machine)).
(Unsupervised learning) A method of extracting the essential structure and features existing in the data by analyzing the actual data itself in the absence of sample data (eg, k-means (K-Means), latent Semantic Index (LSI), Topic Model Method (LDA)).
(Semi-supervised learning) A method in which learning is performed using a small number of samples, then actual data is classified to some extent, and the one with high accuracy among the results is re-learned as a sample. (Example: Bootstrap method, Adaboost)
(Structure learning) A method of collectively estimating individual data in a form optimized for the structure of the entire data, without estimating the data individually. (Eg structured SVM, conditional random field (CRF)).
(Reinforcement learning) There is no sample, but instead, it is used as a clue for further learning by obtaining feedback information after learning. (Example: Bandit algorithm, UCB algorithm).
(Deep learning) A technique that multi-layers a technique called a neural network that mimics the structure of the human brain in a software manner to improve its performance (eg, restricted Boltzmann machine, Category2Vec). FIG. 16 shows a machine learning unit 400 including, for example, a neural network.

The learning method for machine learning may be any of supervised learning, unsupervised learning, semi-supervised learning, structure learning, reinforcement learning, and deep learning, and a learning method combining these learning methods. However, the learning method for machine learning does not matter.

When the replacement condition proposing unit 4 is the machine learning unit 400 that is machine-learned using the corrected sentence and the uncorrected sentence, the replacement error may occur regardless of the field of the analysis target sentence, the replacement condition, or the like. It is possible to determine a replacement condition with less.

The substitution condition proposing unit 41 uses a character described in a fixed range of the corrected sentence and a character described in the same range as the fixed range in the uncorrected sentence to perform a machine learning machine learning unit. When it is 400, the optimum replacement condition can be promptly proposed.

1: Information processing apparatus 1A: Information processing system 2: Image data generation unit 3: Character replacement unit 4: Replacement condition proposal unit 5: Error character notification unit 6: Correction acceptance unit 10: First storage unit 20: Second storage unit 41 : Replacement condition proposing unit 100A: Screen 100B: Screen 101: Processor 102: Memory 103: Input/output interface 104: Bus 200: Communication network 300: Server 400: Machine learning unit

JP, 2014-081753, A

Claims (9)

An image data generation unit that generates image data of the second character based on a character image showing an image of a second character having a character shape different from the character shape of the first character described in the sentence;
In the sentence, when a character positioned before or after the first character matches a specific replacement condition, a character replacement unit that replaces the first character with the second character based on image data of the second character,
An information processing apparatus including. The information processing apparatus according to claim 1, further comprising a correction acceptance unit that accepts a correction by the user for the sentence including the second character replaced by the character replacement unit. When the correction receiving unit receives a correction made to the sentence by the user, the plurality of replacement conditions used for the replacement with the second character by comparing the corrected sentence with the uncorrected sentence. The information processing apparatus according to claim 2, further comprising: a replacement condition suggesting unit configured to analyze a correlation between the replaced second character and the replaced second character, and notify the replacement condition with few replacement errors by a display message or a voice message. . The information processing apparatus according to claim 3, wherein the replacement condition proposing unit is a machine learning unit that is machine-learned using the corrected sentence and the uncorrected sentence. When the correction accepting unit receives a correction by the user for a certain range of the sentence, the characters described in the certain range of the corrected sentence and the same range of the uncorrected sentence as the certain range are described. By analyzing the correlation between the content of the text within the certain range before correction and the content of the text within the certain range after modification by comparing the characters that have been corrected, and the replacement condition for other than the certain range of the text. The information processing apparatus according to claim 2, further comprising: a replacement condition proposing unit that notifies, by a display message or a voice message. The replacement condition proposing unit uses characters described in a certain range of the corrected sentence and characters described in the same range as the certain range in the sentence in which the first character is described, The information processing apparatus according to claim 5, which is a machine learning unit that has been machine-learned. Based on the typographical error determination condition for determining that the character described in the sentence before being replaced with the second character is erroneous, it is detected that the character matching the substitution condition is typographical error, and the detected result is displayed. The information processing apparatus according to any one of claims 3 to 6, further comprising a typographical error notifying unit that notifies a message or a voice message. An information processing system comprising: the information processing apparatus according to claim 1; a communication network; and a processing apparatus communicatively connected to the information processing apparatus via the communication network. On the computer,
Generating image data of the second character based on a character image showing an image of a second character having a character shape different from the character shape of the first character described in the sentence;
An information processing program for replacing the first character with the second character based on the image data of the second character when a character positioned before or after the first character in the sentence matches a specific replacement condition.

Images