JP6613942B2 - Information output system, information output method, and information output program - Google Patents

Description

  The present invention relates to an information output system, an information output method, and an information output program.

  In a bookstore, for example, when creating a planning shelf in which books of a specific genre are arranged on a part of a book shelf, the books are selected based on the know-how of the bookstore staff. In recent years, it has also been proposed to purchase and arrange books based on sales at online bookstores, search results, and the like. When these books are selected, for example, a bookstore clerk performs a book selection work using a keyword search.

JP 2009-271795 A JP 2012-168925 A

  However, in a simple keyword search, only books that match the keyword are searched. In addition, the selection of keywords to be searched is also biased according to the knowledge of the bookstore worker who performs the work and the field of expertise. For this reason, for example, when searching for a book according to a certain plan, it is difficult to output a book that is surprising to the plan as search result data.

  In one aspect, the present invention provides an information output system, an information output method, and an information output program that can enhance the unexpectedness of output data.

  In one aspect, the information output system includes a reception unit, a setting unit, an extraction unit, and an output control unit. The accepting unit accepts input of a first keyword to be processed and input of information specifying a range to be included in the second keyword group to be processed together with the first keyword. The setting unit refers to a first storage unit that stores related word data indicating a hierarchical relationship between the keywords in association with a value indicating the closeness of the relationship between the two keywords. As a starting point, keyword groups existing up to the hierarchy specified by the information specifying the range are extracted and set as the second keyword group. The extraction unit refers to a second storage unit that stores, as summary information, information related to the content for each content, and information on content having a degree of match with the first keyword and the second keyword group of a predetermined value or more To extract. The output control unit outputs a list of the extracted content information.

  The unexpectedness of the output data can be increased.

FIG. 1 is a block diagram illustrating an example of the configuration of the information output system according to the embodiment. FIG. 2 is a diagram illustrating an example of comparison between automatic selection and manual selection. FIG. 3 is a diagram illustrating an example of a related word dictionary storage unit. FIG. 4 is a diagram illustrating an example of the bibliographic information storage unit. FIG. 5 is a diagram illustrating an example of the word information storage unit. FIG. 6 is a diagram illustrating an example of the sales information storage unit. FIG. 7 is a diagram illustrating an example of the bibliographic master storage unit. FIG. 8 is a diagram illustrating an example of hierarchical access. FIG. 9 is a diagram illustrating an example of an output result. FIG. 10 is a diagram illustrating an example of the input screen. FIG. 11 is a diagram illustrating an example of an output screen. FIG. 12 is a flowchart illustrating an example of word information generation processing according to the embodiment. FIG. 13 is a flowchart illustrating an example of the information output process according to the embodiment. FIG. 14 is a diagram illustrating an example of a computer that executes an information output program.

  Hereinafter, embodiments of an information output system, an information output method, and an information output program disclosed in the present application will be described in detail based on the drawings. The disclosed technology is not limited by the present embodiment. Further, the following embodiments may be appropriately combined within a consistent range.

  FIG. 1 is a block diagram illustrating an example of the configuration of the information output system according to the embodiment. An information output system 1 illustrated in FIG. 1 includes a terminal device 10 and a server 100. 1 shows an example in which the number of terminal devices 10 is one, the number of terminal devices 10 is not limited, and an arbitrary number of terminal devices 10 may be included. In FIG. 1, the server 100 is a single device, but each function unit and each storage unit of the server 100 may be configured using a cloud computing service such as SaaS (Software as a Service). . In this embodiment, description will be made using books as an example of content.

  The terminal device 10 and the server 100 are connected via a network N so that they can communicate with each other. As such a network N, any type of communication network such as the Internet, a LAN (Local Area Network), and a VPN (Virtual Private Network) can be adopted regardless of wired or wireless.

  The terminal device 10 is, for example, a terminal device used by a store clerk (hereinafter also referred to as an operator), and a keyword and importance level information used for selection are input on an input screen displayed on a display unit (not shown). . The terminal device 10 transmits the input keyword and importance level information to the server 100 via the network N as the first keyword and importance level information. In addition, the terminal device 10 receives an output screen including the selection result from the server 100 via the network N. The terminal device 10 displays the received output screen on a display unit (not shown).

  Here, as a 1st keyword, what is necessary is just the word group relevant to a selection theme, for example, and the outline of the book which the operator picked up may be input besides a word. The importance level information includes, for example, items such as unexpectedness (serendipity), sales importance level, author popularity importance level, target genre, and target publication year. The unexpectedness, the sales importance level, and the author popularity importance level can be selected from, for example, three levels of high, medium, and low.

  Here, the unexpectedness is an item indicating, for example, whether it is easy to recall from a keyword, that is, whether a keyword directly related to the keyword is included. The sales importance level is an item indicating, for example, whether or not to place importance on the sales amount or the number of sales. The author popularity emphasis level is an item indicating, for example, whether or not to place importance on the sales amount or the number of sales of books by the author. When the unexpectedness is designated as “high”, the search process is executed so that even a book including a keyword that is not directly related to the keyword is easily selected as compared with the case where “unusual” is designated. When the sales importance level is designated as “high”, the search processing is executed so that a book with a larger sales amount or number of sales is more easily selected than when it is designated as “low”. If the author ’s popularity is specified as “high”, the higher the book sales or number of books by that author, the easier it is to select. Execute the process.

  In addition, as the target genre, a genre such as a novel or a criticism can be selected. Furthermore, the target publication year can be input in the Western calendar, for example.

  As an example of the terminal device 10, a tablet terminal can be adopted. As the terminal device 10, not only a portable terminal such as the above-described tablet terminal but also a stationary personal computer can be adopted as the terminal device 10. In addition to the above-described tablet terminals, the terminal device 10 may employ, for example, a portable personal computer, a mobile communication terminal such as a smartphone or a mobile phone, and the like as the portable terminal.

  The server 100 receives the first keyword and the importance level information from the terminal device 10. The first keyword is an example of a first keyword to be processed. The importance level information is an example of information for specifying a range to be included in the second keyword group to be processed together with the first keyword. The server 100 receives input of a first keyword to be processed and input of information specifying a range to be included in the second keyword group to be processed together with the first keyword. The server 100 refers to the first storage unit that stores the related word data indicating the hierarchical relationship between the keywords in association with the value indicating the closeness of the relationship between the two keywords, and starts from the first keyword. As described above, a keyword group up to a hierarchy specified by the information specifying the range is extracted and set as the second keyword group. The server 100 refers to a second storage unit that stores content-related information as summary information for each content, and extracts content information whose degree of match with the first keyword and the second keyword group is greater than or equal to a predetermined value. . The server 100 transmits a list of extracted content information, for example, data of a selection result to the terminal device 10 as an output screen and outputs it. Thereby, the information output system 1 can improve the unexpectedness of the output data. In other words, it is possible to increase the possibility of including in the search results information that is likely to be surprising to the operator. And if the books shown in such search results are placed in the book planning corner, it will be a planning corner that is surprising and interesting to the bookstore users. it can.

  In the present embodiment, in addition to the classification based on the genre that has been used in the past, the reason for purchase of the consumer is captured using words such as mood, values, and background at the time of purchase. For example, conventionally used genres include “comics”, “non-fiction”, “fiction”, “practical books”, and the like. Examples of words related to mood at the time of purchase include “when you want to laugh” and “when you want to cry”. Examples of values related to values include “outdoor” and “indoor”. Examples of background words include “when giving birth” and “when getting married”. Further, in this embodiment, the existing purchase genre is also subdivided to capture the consumer's reason for purchase in more detail.

  Next, a comparison between automatic selection and manual selection will be described with reference to FIG. FIG. 2 is a diagram illustrating an example of comparison between automatic selection and manual selection. As shown in FIG. 2, as for the number of books to be selected, automatic selection can be selected using all books as a parameter, whereas manual selection requires a work amount corresponding to the number of selected books. Therefore, it becomes a selection book with the number of books corresponding to the amount of work as a parameter. The bias of genre, etc., can be set by parameters (importance level information) that prioritize automatic selection, whereas manual selection depends on the knowledge of the operator, such as genre and publishing age. Bias occurs. In addition, the degree of consideration of sales and the importance of author popularity can be factored into the logic using automatic selection, whereas manual selection requires information such as sales if it is sales. Reference is required.

  Next, the configuration of the server 100 will be described. As illustrated in FIG. 1, the server 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The server 100 may include various functional units included in known computers, for example, functional units such as various input devices and audio output devices, in addition to the functional units illustrated in FIG.

  The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is a communication interface that is connected to the terminal device 10 via the network N in a wired or wireless manner and manages communication of information with the terminal device 10. The communication unit 110 receives the first keyword and the importance level information from the terminal device 10. The communication unit 110 outputs the received first keyword and importance level information to the control unit 130. Further, the communication unit 110 transmits the input screen, the output screen, and the like input from the control unit 130 to the terminal device 10.

  The storage unit 120 is realized by a storage device such as a RAM (Random Access Memory), a semiconductor memory element such as a flash memory, a hard disk, or an optical disk, for example. The storage unit 120 includes a related word dictionary storage unit 121, a bibliographic information storage unit 122, a word information storage unit 123, a sales information storage unit 124, and a bibliographic master storage unit 125. In addition, the storage unit 120 stores information used for processing in the control unit 130.

  The related word dictionary storage unit 121 stores related word data. The related word dictionary storage unit 121 is an example of a first storage unit. The related term data is data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of the relationship between two keywords. Note that the hierarchical relationship means that hierarchical access is possible when referring to the related word dictionary storage unit 121. FIG. 3 is a diagram illustrating an example of a related word dictionary storage unit. As shown in FIG. 3, the related word dictionary storage unit 121 is a numerical value that represents, as related word data, the keywords on the vertical axis and the horizontal axis, and the closeness of the relationship between the keyword on the vertical axis and the keyword on the horizontal axis. The similarity between words is memorized.

  In the example of FIG. 3, for each of “facts”, “past”, “startle”, and “shock” in each column with respect to “truth” in the first row, for example, the similarity between words is 0 to 1 It is expressed as a range. For example, “truth” and “facts” indicate that the similarity between words is “0.91”. The keywords in each column are sorted in descending order of similarity between words.

  Returning to the description of FIG. 1, the bibliographic information storage unit 122 stores a book name and a synopsis of the book in association with each other. The bibliographic information storage unit 122 is an example of a second storage unit. FIG. 4 is a diagram illustrating an example of the bibliographic information storage unit. As shown in FIG. 4, the bibliographic information storage unit 122 has items such as “book name” and “summary”. The bibliographic information storage unit 122 stores, for example, one record for each book name.

  “Book name” is information indicating the name of the book. “Synopsis” is information indicating a synopsis of a book, that is, bibliographic information. Bibliographic information is an example of summary information. In the example of the first line in FIG. 4, the book name “Forensic Investigation Group xxx” and the synopsis “The Truth of Surprise Reviving from the Past.

  Returning to the description of FIG. 1, the word information storage unit 123 stores vectorized word information. The word information storage unit 123 is an example of a third storage unit. FIG. 5 is a diagram illustrating an example of the word information storage unit. As illustrated in FIG. 5, the word information storage unit 123 includes items such as “word”, “v1”, “v2”, “v3”... “V100”. The word information storage unit 123 stores, for example, one record for each word.

  “Word” is bibliographic information, that is, information indicating a word extracted from a synopsis or the like. “V1”, “v2”, “v3”... “V100” are vectorized word information. The number after “v” in the vectorized word information is a numerical value representing the axis of the vector. For vectorization of words, for example, the technique of the document “Mikolov et al .: Distributed Representations of Words and Phrases and their Compositionality (2013)” can be used.

  Returning to the description of FIG. 1, the sales information storage unit 124 stores the number of sales volumes for each book, for example, for each month. FIG. 6 is a diagram illustrating an example of the sales information storage unit. As illustrated in FIG. 6, the sales information storage unit 124 includes items such as “book name” and “number of sales books”. The sales information storage unit 124 stores, for example, one record for each book name.

  “Book name” is information indicating the name of the book. The “number of sales volumes” is information indicating the number of volumes sold every month, for example. In the example of the first line in FIG. 6, the book name “Forensic Investigation Group xxx” sold “258” in January 2016 and sold “357” in February 2016.

  Returning to the description of FIG. 1, the bibliographic master storage unit 125 stores the book name, the publication year, the author, and the genre of the book in association with each other. FIG. 7 is a diagram illustrating an example of the bibliographic master storage unit. As illustrated in FIG. 7, the bibliographic master storage unit 125 includes items such as “book name”, “publishing year”, “author”, and “genre”. The bibliographic master storage unit 125 stores, for example, one record for each book name.

  “Book name” is information indicating the name of the book. “Publishing year” is information indicating the year in which the book was published. “Author” is information indicating the author of the book. “Genre” is information indicating the genre of a book such as a novel and a criticism, for example. In the example of the first line in FIG. 7, the book name “Forensic Investigation Group xxx” is published in “2005”, the author is “Tanaka XX”, and the genre is “criticism”.

  Returning to the description of FIG. 1, the control unit 130 executes, for example, a program stored in an internal storage device using a RAM as a work area by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. Is realized. Further, the control unit 130 may be realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The control unit 130 includes a reception unit 131, a calculation unit 132, a setting unit 133, a priority calculation unit 134, an extraction unit 135, and an output control unit 136. Information processing functions and Realize or execute the action. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 1, and may be another configuration as long as the information processing described below is performed.

  The accepting unit 131 transmits an input screen to the terminal device 10 via the communication unit 110 and the network N. The first keyword and importance level information input from the terminal device 10 on the input screen are input to the reception unit 131 via the communication unit 110 and the network N. The accepting unit 131 accepts the input first keyword as a processing target. The accepting unit 131 accepts the input importance level information as information for specifying a range to be included in the second keyword group to be processed together with the first keyword. In the following description, information for specifying a range is also referred to as importance level information. The accepting unit 131 outputs the accepted first keyword to the calculating unit 132 and the setting unit 133. In addition, the reception unit 131 outputs the received importance information to the setting unit 133 and the priority calculation unit 134.

  The calculation unit 132 once executes a word information generation process for vectorizing a word included in the bibliographic information and storing it in the word information storage unit 123 for the bibliographic information of the book stored in the bibliographic information storage unit 122. That is, the word information generation process generates word information that is the basis of a related word dictionary used when performing automatic selection based on bibliographic information of all books to be searched.

  The calculation unit 132 acquires bibliographic information of all books from the bibliographic information storage unit 122 as word information generation processing. The calculation unit 132 vectorizes words included in the acquired bibliographic information. The calculation unit 132 stores the word and the vector in the word information storage unit 123 in association with each other. That is, the calculation unit 132 stores the vectorized word information in the word information storage unit 123.

  When the first keyword is input from the accepting unit 131, the calculating unit 132 refers to the word information storage unit 123 and reads the value of the vectorized word information corresponding to the first keyword. The calculation unit 132 calculates the similarity between words based on the value of the read vectorized word information. In other words, the calculation unit 132 refers to the word information storage unit 123 that stores a value indicating the degree of a plurality of predetermined items for each keyword, and stores each of the plurality of keywords including the first keyword in association with each other. Read the corresponding value. Further, the calculation unit 132 calculates a value indicating the closeness of the relationship between the keywords including the first keyword based on the read value. Here, the value indicating the degree of a plurality of predetermined items for each keyword corresponds to the value of the vectorized word information, and the value indicating the closeness of the relationship between the keywords corresponds to the similarity between words. .

  The calculation unit 132 generates related word data based on the calculated similarity between words. That is, the calculation unit 132 generates related word data corresponding to the first keyword. The calculation unit 132 stores the generated related word data in the related word dictionary storage unit 121. This is because if the related word data is generated for all the words included in the bibliographic information, the related word data becomes enormous. For this reason, the calculation unit 132 generates partial related word data based on the vectorized word information corresponding to the first keyword used for processing.

  In other words, the calculation unit 132 generates a hierarchical relationship with other keywords having a relationship with the first keyword, based on the calculated similarity between words. That is, the calculation unit 132 is an example of a generation unit that generates a hierarchical relationship with another keyword having a relationship with the first keyword based on a value indicating the calculated relationship. The calculation unit 132 may output the generated related word data to the output control unit 136, and the output control unit 136 may store the related word data in the related word dictionary storage unit 121. In addition, when the calculation unit 132 stores the generated related word data in the related word dictionary storage unit 121, the calculation unit 132 outputs a setting instruction to the setting unit 133.

  The setting unit 133 receives the first keyword and the importance level information from the reception unit 131. When a setting instruction is input from the calculation unit 132, the setting unit 133 refers to the related word dictionary storage unit 121, and specifies based on the unexpectedness in the importance level information, starting from the input first keyword. A keyword group up to the hierarchy to be extracted is extracted. Note that the unexpectedness can be, for example, three stages of high, medium and low. Hierarchical access for unexpectedness is, for example, “width: 5 and depth: 5” if the unexpectedness is high, and “width: 4 and depth: 3” if the unexpectedness is moderate. Can be “width: 3, depth: 1”. That is, when the unexpectedness is high, many related words are extracted, and when the unexpectedness is low, few related words are extracted. Note that the unexpectedness may be specified directly by the operator in width and depth.

  The setting unit 133 sets the extracted keyword group as the second keyword group. That is, the setting unit 133 sets a second keyword group having a hierarchical relationship with the first keyword. The setting unit 133 outputs the first keyword and the second keyword group to the extraction unit 135.

  Here, hierarchical access using a hierarchical relationship will be described with reference to FIG. FIG. 8 is a diagram illustrating an example of hierarchical access. In the example of FIG. 8, “truth” is input as the first keyword, and the layer based on the importance level information is “depth: 2, first layer width: 3, second layer width: 2”. It shows the access range in. For example, the setting unit 133 refers to the related word dictionary storage unit 121 illustrated in FIG. 3, and, in descending order of similarity between words with respect to “truth”, “facts” and “ "Past" and "Amazing" are extracted.

  Furthermore, the setting unit 133 extracts words corresponding to “width of the second layer: 2” in descending order of similarity between words with respect to the extracted “facts”, “past”, and “amazing”. The setting unit 133 extracts “reason” and “memory” in descending order of similarity between words with respect to “facts”. The setting unit 133 extracts “surprise” and “impact” in descending order of similarity between words with respect to “amazing”. The setting unit 133 extracts “disappointing” and “truth” in descending order of similarity between words with respect to “past”. The setting unit 133 extracts the extracted words in the first and second layers, that is, “facts”, “past”, “surprise”, “reason”, “memory”, “surprise”, “shock”, “disappointing”. , “Truth” is set as the second keyword group.

  Returning to the description of FIG. 1, when the priority level information is input from the reception unit 131, the priority level calculation unit 134 refers to the sales information storage unit 124 and the bibliographic master storage unit 125. The priority for each book is calculated based on information such as importance, author popularity importance, target genre, and target publication year. That is, the priority calculation unit 134 calculates priorities for all books stored in the sales information storage unit 124 and the bibliography master storage unit 125. For example, when the priority is based on sales, the number of sales volumes of all books is graphed, the graph is further linearized, and the priority is calculated based on the slope of the straight line. In the calculation of the priority, the slope of the straight line is increased when the importance of sales is increased, and the slope of the straight line is decreased when the importance is decreased. The priority is represented by a numerical value in the range of 0 to 1, for example. The priority calculation unit 134 outputs the calculated priority for each book to the extraction unit 135.

  When the first keyword and the second keyword group are input from the setting unit 133, the extraction unit 135 refers to the bibliographic information storage unit 122, and determines the degree of match between the first keyword and the second keyword group. Calculate for each book. In other words, the extraction unit 135 calculates the degree of match for all books stored in the bibliographic information storage unit 122. In other words, the extraction unit 135 uses a keyword group obtained by adding the second keyword group to the first keyword as a search keyword. The degree of match is represented by a numerical value in the range of 0 to 1, for example.

  In addition, when the priority for each book is input from the priority calculation unit 134, the extraction unit 135 calculates the extraction score by adding the matching degree and the priority for each book. When the extraction unit 135 calculates the extraction score, the extraction unit 135 refers to the bibliographic information storage unit 122 and extracts information on a book whose calculated extraction score is a predetermined value or more. Here, the extraction score being equal to or greater than a predetermined value can be, for example, a value from which the top three extraction scores are extracted. Note that the book information includes, for example, a book name, a synopsis, and an extraction score. Note that the extraction score may exceed 1, and it is only necessary to be able to sort in descending order or ascending order. The extraction unit 135 outputs the extracted book information to the output control unit 136.

  Note that the first keyword may be a first keyword group including a plurality of keywords. In this case, the extraction unit 135 switches the method of extracting information on books with a matching degree equal to or greater than a predetermined value according to the number of keywords indicated in the first keyword group and the second keyword group. For example, when the number of keywords is large, the extraction unit 135 can perform extraction using a naive Bayes model, and when the number of keywords is small, the extraction unit 135 can perform extraction using a distance between sentences. Further, the extraction unit 135 may extract information on books having a matching degree equal to or higher than a predetermined value in each of a plurality of types of extraction methods. Note that the above-described extraction score may be used as the degree of match in these cases. Further, the degree of match is equal to or greater than a predetermined value, for example, can be a value from which the top three match degrees are extracted. When the extraction method of the book information is switched, the extraction unit 135 may output the algorithm used for the extraction to the output control unit 136 together with the extracted book information to be displayed on the output screen.

  When the book information extracted from the extraction unit 135 is input, the output control unit 136 generates an output screen including the input book information, that is, a selection result. The output control unit 136 transmits the generated output screen to the terminal device 10 via the communication unit 110 and the network N, and displays the output screen on a display unit (not shown) of the terminal device 10. Further, when the extraction unit 135 extracts information on a book by a plurality of types of extraction methods, the output control unit 136 displays an output screen by combining the extraction results of each of the plurality of types of extraction methods. Generate.

  Here, the output result will be described with reference to FIG. FIG. 9 is a diagram illustrating an example of an output result. FIG. 9 is an example of an output result when “truth” is input as the first keyword. The first line in FIG. 9, that is, the book with the highest score is “Forensic Investigation Group xxx”, and the summary is “truth” as the first keyword and “past” in the second keyword group. “Surprised” is included. Also, the book “Chinese Economy xxx” on the second line in FIG. 9 includes “impact” and “facts” in the second keyword group. In other words, the book “Chinese Economy xxx” does not include the first keyword “truth”, but includes a word belonging to the second keyword group, and thus is extracted, that is, selected.

  As another example, for example, when “simple” and “tidy” are input as the first keyword, the output result of the selection is not only a practical book related to cleaning, Books related to yoga and fitness are selected from other genres related to children's books. That is, in this embodiment, the selection result output includes unexpectedness. If the book shown in the output result is placed in the book planning corner, it will be surprising to the bookstore users and attracted interest. It can be a planning corner where you can be interested in and expect to buy.

  Subsequently, the input screen and the output screen will be described with reference to FIGS. 10 and 11. FIG. 10 is a diagram illustrating an example of the input screen. An input screen 30 shown in FIG. 10 is displayed on a display unit (not shown) of the terminal device 10, and includes an input field 31 for inputting a word or book name to be searched, and an input field 32 for inputting importance level information. And a selection execution button 33 for executing selection. On the input screen 30, for example, a word is input to the input field 31 by an operator, and after the check box of each item in the input field 32 is checked, the selection execution button 33 is pressed. When the selection execution button 33 is pressed, the terminal device 10 transmits the word in the input field 31 as the first keyword and the input content in the input field 32 as importance level information to the server 100.

  FIG. 11 is a diagram illustrating an example of an output screen. An output screen 40 shown in FIG. 11 is displayed on a display unit (not shown) of the terminal device 10, and an input field 41 that displays the word or book name used for the search in an editable manner and the importance level information used for the search can be edited. And an input field 42 to be displayed. The output screen 40 also includes a related word column 43 that displays the related words used in the selection, and a selection re-execution that executes the selection again when the contents are edited in the input column 41, the input column 42, and the related word column 43. And an execution button 44. The output screen 40 has an output result column 45 that displays the output result of the selection. The contents input on the input screen 30 are displayed in the input field 41 and the input field 42. For example, the operator may exclude the output result of the selection in the output result field 45 by changing the check box of each item in the input field 42 or by clicking the related word in the related word field 43. Then, the conditions can be changed and the selection book can be executed again.

  Next, the operation of the information output system 1 of the embodiment will be described. First, the word information generation process will be described. FIG. 12 is a flowchart illustrating an example of word information generation processing according to the embodiment.

  For example, when the calculation unit 132 of the server 100 receives an instruction from the administrator of the server 100 to execute the word information generation process, the calculation unit 132 executes the word information generation process once. The calculation unit 132 acquires bibliographic information of all books from the bibliographic information storage unit 122 (step S1). The calculation unit 132 vectorizes words included in the acquired bibliographic information (step S2). The calculation unit 132 stores the word and the vector in association with each other in the word information storage unit 123 (step S3). As a result, the server 100 can generate word information that is the basis of the related word dictionary used when the automatic selection is performed.

  Next, the information output process will be described. FIG. 13 is a flowchart illustrating an example of the information output process according to the embodiment.

  The reception unit 131 of the server 100 transmits an input screen to the terminal device 10. When the terminal device 10 receives the input screen, the terminal device 10 displays the received input screen on a display unit (not shown), and accepts input of keywords and importance level information from the operator. The terminal device 10 transmits the input keyword and importance level information to the server 100 as the first keyword and importance level information. The receiving unit 131 of the server 100 receives the first keyword received and input and the importance level information as processing targets (step S11). The accepting unit 131 outputs the accepted first keyword to the calculating unit 132 and the setting unit 133. In addition, the reception unit 131 outputs the received importance information to the setting unit 133 and the priority calculation unit 134.

  When the first keyword is input from the accepting unit 131, the calculating unit 132 refers to the word information storage unit 123 and calculates the similarity between words corresponding to the first keyword (step S12). The calculation unit 132 generates related word data based on the calculated similarity between words. The calculation unit 132 stores the generated related word data in the related word dictionary storage unit 121 (step S13). When the calculation unit 132 stores the generated related word data in the related word dictionary storage unit 121, the calculation unit 132 outputs a setting instruction to the setting unit 133.

  When a setting instruction is input from the calculation unit 132, the setting unit 133 refers to the related word dictionary storage unit 121, and specifies based on the unexpectedness in the importance level information, starting from the input first keyword. A keyword group up to the hierarchy to be extracted is extracted. The setting unit 133 sets the extracted keyword group as the second keyword group (step S14). The setting unit 133 outputs the first keyword and the second keyword group to the extraction unit 135.

  When the first keyword and the second keyword group are input from the setting unit 133, the extraction unit 135 refers to the bibliographic information storage unit 122, and determines the degree of match between the first keyword and the second keyword group. Calculation is made for each book (step S15).

  When priority level information is input from the reception unit 131, the priority level calculation unit 134 refers to the sales information storage unit 124 and the bibliographic master storage unit 125, and calculates a priority level for each book based on the priority level information. (Step S16). The priority calculation unit 134 outputs the calculated priority for each book to the extraction unit 135.

  When the priority for each book is input from the priority calculation unit 134, the extraction unit 135 calculates an extraction score based on the degree of matching and the priority for each book (step S17). When the extraction unit 135 calculates the extraction score, the extraction unit 135 refers to the bibliographic information storage unit 122 and extracts information on a book whose calculated extraction score is equal to or greater than a predetermined value (step S18). The extraction unit 135 outputs the extracted book information to the output control unit 136.

  When the book information extracted from the extraction unit 135 is input, the output control unit 136 generates an output screen including the input book information. The output control unit 136 transmits the generated output screen to the terminal device 10 and outputs it, and displays it on a display unit (not shown) of the terminal device 10 (step S19). Thereby, the information output system 1 can improve the unexpectedness of the output data. That is, the information output system 1 can generate a related word dictionary from all the synopsis information and select a book including related words, so that a book corresponding to a theme with higher ambiguity can be selected. In addition, since the information output system 1 accesses hierarchically based on the width and depth specified in the related word dictionary, the degree of unexpectedness in the output result of the selection is controlled by changing the reference level of the related word. it can. Furthermore, since the information output system 1 can switch the algorithm for calculating the degree of matching according to the number of input words, and can add the priority and the degree of matching, it can cope with the unexpectedness of the selection required by the operator. .

  As described above, the information output system 1 accepts the input of the first keyword to be processed and the input of information specifying the range to be included in the second keyword group to be processed together with the first keyword. In addition, the information output system 1 refers to a first storage unit that stores related word data that indicates a hierarchical relationship between keywords in association with a value that indicates the closeness of the relationship between two keywords. Starting from the keyword, the keyword group existing up to the hierarchy specified by the information specifying the range is extracted and set as the second keyword group. In addition, the information output system 1 refers to the second storage unit that stores content information for each content as summary information, and the degree of match between the first keyword and the second keyword group is greater than or equal to a predetermined value. Extract information. Also, the information output system 1 outputs a list of extracted content information. As a result, the unexpectedness of the output data can be enhanced.

  In the information output system 1, the content is a book. In addition, the information output system 1 refers to a second storage unit that stores information related to the book as bibliographic information for each book, and the degree of coincidence between the first keyword and the second keyword group is greater than or equal to a predetermined value. Extract information. As a result, the unexpectedness of the data output as the selection output result can be enhanced.

  Further, when the information output system 1 further receives an input of the first keyword, the information output system 1 refers to a third storage unit that stores a value indicating the degree of a plurality of predetermined items for each keyword, and determines the first keyword. A value stored in association with each of the plurality of included keywords is read, and a value indicating the closeness of the relationship between the keywords including the first keyword is calculated based on the read value. Further, the information output system 1 generates a hierarchical relationship with other keywords having a relationship with the first keyword, based on the value indicating the closeness of the calculated relationship. Further, the information output system 1 outputs the generated hierarchical relationship to the first storage unit as related word data. As a result, since the related word dictionary based on the minimum related word data is referred to, the processing time of the book selection can be shortened.

  In the information output system 1, the first keyword is a first keyword group including a plurality of keywords. In addition, the information output system 1 switches a method of extracting information on content having a matching degree equal to or higher than a predetermined value according to the number of keywords indicated in the first keyword group and the second keyword group. As a result, an appropriate extraction method can be taken according to the number of keywords.

  In addition, the information output system 1 extracts information on content having a matching degree equal to or greater than a predetermined value in each of a plurality of types of extraction methods. The information output system 1 also outputs the results of extraction by each of a plurality of types of extraction methods. As a result, an output result corresponding to the extraction method can be presented to the worker.

  In addition, the information output system 1 has a score obtained by adding the priority for each content calculated based on at least one of the sales importance level, the author popularity importance level, the genre, and the date to the degree of match. Extract content information for. As a result, it is possible to set how much to refer to these items.

  Moreover, in the said Example, although the salesclerk of the bookstore was mentioned as an operator, it is not limited to this. For example, a publishing agency that distributes books to each bookstore may be used when proposing a book to the bookstore.

  Moreover, in the said Example, although the book was mentioned and demonstrated as an example of a content, a content is not limited to this. For example, the content may be a product other than a book, or may be applied to a Web site or call center log or Q & A search. In the case of a Web site, a Web article may be used as summary information as it is. As the author popularity, the number of follow readers for the article writer of the website may be used. As sales, the number of PV (Page View) may be used. As the date, the date of publication of the web page can be used. That is, in the case of a Web site, tagging for Web articles and products, classification with a new axis, and the like can be performed. In the case of a call center, it is possible to use inquiry information and summary information based on Q & A, and use importance count information such as the number of accesses, the date of inquiry, and the like. That is, in the case of a call center, it is possible to narrow down inquiry information and Q & A to be referred to by performing a search with a keyword having a high degree of ambiguity.

  In addition, each component of each part illustrated does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each unit is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed / integrated in arbitrary units according to various loads or usage conditions. Can be configured. For example, the reception unit 131 and the calculation unit 132 may be integrated. For example, all or part of the data stored in the storage unit 120 may be stored in the storage unit of another computer connected to the server 100 so as to be communicable. In addition, the illustrated processes are not limited to the above-described order, and may be performed at the same time as long as the process contents are not contradictory, or may be performed in a different order.

  Furthermore, various processing functions performed by each device may be executed entirely or arbitrarily on a CPU (or a microcomputer such as an MPU or MCU (Micro Controller Unit)). In addition, various processing functions may be executed in whole or in any part on a program that is analyzed and executed by a CPU (or a microcomputer such as an MPU or MCU) or on hardware based on wired logic. Needless to say, it is good.

  By the way, the various processes described in the above embodiments can be realized by executing a prepared program on a computer. Therefore, in the following, an example of a computer that executes a program having the same function as each of the above embodiments will be described. FIG. 14 is a diagram illustrating an example of a computer that executes an information output program.

  As illustrated in FIG. 14, the computer 200 includes a CPU 201 that executes various arithmetic processes, an input device 202 that receives data input, and a monitor 203. The computer 200 also includes a medium reading device 204 that reads a program and the like from a storage medium, an interface device 205 for connecting to various devices, and a communication device 206 for connecting to other information processing devices and the like by wire or wirelessly. Have The computer 200 also includes a RAM 207 that temporarily stores various types of information and a hard disk device 208. Each device 201 to 208 is connected to a bus 209.

  The hard disk device 208 includes an information output program having the same functions as the processing units of the reception unit 131, the calculation unit 132, the setting unit 133, the priority calculation unit 134, the extraction unit 135, and the output control unit 136 illustrated in FIG. Is memorized. The hard disk device 208 also includes a related word dictionary storage unit 121, a bibliographic information storage unit 122, a word information storage unit 123, a sales information storage unit 124, a bibliographic master storage unit 125, and various types of information output programs for realizing the information output program. Data is stored. The input device 202 receives input of various information such as operation information from the administrator of the computer 200, for example. The monitor 203 displays various screens such as an output screen for the administrator of the computer 200, for example. The interface device 205 is connected to, for example, a printing device. The communication device 206 has the same function as the communication unit 110 illustrated in FIG. 1, for example, is connected to the network N, and exchanges various types of information with the terminal device 10.

  The CPU 201 reads out each program stored in the hard disk device 208, develops it in the RAM 207, and executes it to perform various processes. In addition, these programs can cause the computer 200 to function as the reception unit 131, the calculation unit 132, the setting unit 133, the priority calculation unit 134, the extraction unit 135, and the output control unit 136 illustrated in FIG.

  Note that the information output program does not necessarily have to be stored in the hard disk device 208. For example, the computer 200 may read and execute a program stored in a storage medium readable by the computer 200. The storage medium readable by the computer 200 corresponds to, for example, a portable recording medium such as a CD-ROM, a DVD disk, a USB (Universal Serial Bus) memory, a semiconductor memory such as a flash memory, a hard disk drive, and the like. Alternatively, the information output program may be stored in a device connected to a public line, the Internet, a LAN, or the like, and the computer 200 may read and execute the information output program therefrom.

  As described above, the following supplementary notes are further disclosed regarding the embodiment including the present example.

(Additional remark 1) The reception part which receives the input of the 1st keyword made into processing object, and the input of the information which specifies the range included in the 2nd keyword group made into processing object with the said 1st keyword,
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, and A setting unit that extracts a keyword group up to a hierarchy specified by information specifying a range and sets the second keyword group;
Extraction for extracting content information whose degree of coincidence with the first keyword and the second keyword group is greater than or equal to the second storage unit that stores information on the content as summary information for each content And
An output control unit for outputting a list of the extracted content information;
An information output system comprising:

(Appendix 2) The content is a book,
The extraction unit refers to the second storage unit that stores information on the book as bibliographic information for each of the books, and the degree of match with respect to the first keyword and the second keyword group is equal to or greater than a predetermined value. Extract book information,
The information output system according to supplementary note 1, wherein:

(Supplementary note 3) Further, when the input of the first keyword is received, a third storage unit that stores a value indicating the degree of a plurality of predetermined items for each keyword is referred to, and the first keyword is included. A calculation unit that reads the value stored in association with each of a plurality of keywords, and calculates a value indicating the closeness of the relationship between the keywords including the first keyword based on the read value When,
A generating unit that generates a hierarchical relationship between the first keyword and another keyword having a relationship based on the calculated value indicating the proximity of the relationship;
Have
The output control unit outputs the generated hierarchical relationship to the first storage unit as the related word data.
The information output system according to appendix 1 or 2, characterized in that:

(Appendix 4) The first keyword is a first keyword group including a plurality of keywords,
The extraction unit switches a method of extracting information on content having a degree of match of a predetermined value or more according to the number of keywords indicated in the first keyword group and the second keyword group.
The information output system according to any one of appendices 1 to 3, characterized in that:

(Additional remark 5) The said extraction part extracts the information of the content whose said matching degree is more than predetermined | prescribed by each of the method of multiple types of extraction,
The output control unit outputs a result of extraction by each of the plurality of types of extraction methods;
The information output system according to any one of appendices 1 to 4, characterized in that:

(Additional remark 6) The said extraction part adds the priority for every said content calculated based on one or more among sales importance, author popularity importance, a genre, and a date to the said matching degree Extract information on content that is greater than or equal to
The information output system according to any one of appendices 2 to 5, characterized in that:

(Appendix 7) Accepting input of a first keyword to be processed and input of information specifying a range to be included in a second keyword group to be processed together with the first keyword,
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, and Extracting a keyword group up to the hierarchy specified by the information specifying the range and setting it as the second keyword group;
For each content, refer to a second storage unit that stores information related to the content as summary information, and extract information on content that has a degree of match with the first keyword and the second keyword group of a predetermined value or more,
Outputting a list of information of the extracted content;
An information output method, wherein a computer executes a process.

(Appendix 8) The content is a book,
The extracting process refers to the second storage unit that stores information on the book as bibliographic information for each book, and the degree of match between the first keyword and the second keyword group is greater than or equal to a predetermined value. Extract book information for,
The information output method according to appendix 7, wherein

(Additional remark 9) Furthermore, when the input of the first keyword is received, a third storage unit that stores a value indicating the degree of a plurality of predetermined items for each keyword is referred to, and the first keyword is included. Reading the value stored in association with each of a plurality of keywords, based on the read value, to calculate a value indicating the closeness of the relationship between the keywords including the first keyword,
Generating a hierarchical relationship with other keywords having a relationship with the first keyword based on the calculated value indicating the closeness of the relationship;
The computer executes the process,
The output process outputs the generated hierarchical relationship to the first storage unit as the related word data.
The information output method according to appendix 7 or 8, characterized in that:

(Supplementary Note 10) The first keyword is a first keyword group including a plurality of keywords,
The extracting process switches the method of extracting information on the content having the matching degree equal to or greater than a predetermined value according to the number of keywords indicated in the first keyword group and the second keyword group.
The information output method according to any one of appendices 7 to 9, characterized in that:

(Additional remark 11) The said process to extract extracts the information of the content whose said matching degree is more than predetermined | prescribed by each of the method of multiple types of extraction,
The output processing is performed by combining the extraction results of each of the plurality of types of extraction methods.
The information output method according to any one of appendices 7 to 10, characterized in that:

(Additional remark 12) The said process to extract added the priority for every said content calculated based on one or more among sales importance, author popularity importance, a genre, and a date to the said matching degree Extract information about content with a score above a certain level,
The information output method according to any one of appendices 8 to 11, characterized in that:

(Supplementary note 13) Accepting input of a first keyword to be processed and input of information specifying a range to be included in the second keyword group to be processed together with the first keyword,
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, and Extracting a keyword group up to the hierarchy specified by the information specifying the range and setting it as the second keyword group;
For each content, refer to a second storage unit that stores information related to the content as summary information, and extract information on the content having a degree of match with the first keyword and the second keyword group of a predetermined value or more,
Outputting a list of information of the extracted content;
An information output program for causing a computer to execute processing.

(Supplementary Note 14) The content is a book,
The extracting process refers to the second storage unit that stores information on the book as bibliographic information for each book, and the degree of match between the first keyword and the second keyword group is greater than or equal to a predetermined value. Extract book information for,
The information output program according to appendix 13, characterized by:

(Supplementary Note 15) Further, when the input of the first keyword is received, a third storage unit that stores a value indicating the degree of a plurality of predetermined items for each keyword is referred to, and the first keyword is included. Reading the value stored in association with each of a plurality of keywords, based on the read value, to calculate a value indicating the closeness of the relationship between the keywords including the first keyword,
Generating a hierarchical relationship with other keywords having a relationship with the first keyword based on the calculated value indicating the closeness of the relationship;
Let the computer execute the process,
The output process outputs the generated hierarchical relationship to the first storage unit as the related word data.
15. An information output program according to appendix 13 or 14, characterized in that:

(Supplementary Note 16) The first keyword is a first keyword group including a plurality of keywords,
The extracting process switches the method of extracting information on the content having the matching degree equal to or greater than a predetermined value according to the number of keywords indicated in the first keyword group and the second keyword group.
The information output program according to any one of supplementary notes 13 to 15, characterized in that:

(Additional remark 17) The said process to extract extracts the information of the content whose said matching degree is more than predetermined | prescribed by each of the method of multiple types of extraction,
The output processing is performed by combining the extraction results of each of the plurality of types of extraction methods.
The information output program according to any one of supplementary notes 13 to 16, characterized in that:

(Additional remark 18) The said process to extract added the priority for every said content calculated based on one or more among sales importance, author popularity importance, a genre, and a date to the said matching degree Extract information about content with a score above a certain level,
The information output program according to any one of supplementary notes 14 to 17, characterized in that:

DESCRIPTION OF SYMBOLS 1 Information output system 10 Terminal apparatus 100 Server 110 Communication part 120 Storage part 121 Related word dictionary storage part 122 Bibliographic information storage part 123 Word information storage part 124 Sales information storage part 125 Bibliographic master storage part 130 Control part 131 Reception part 132 Calculation part 133 Setting Unit 134 Priority Calculation Unit 135 Extraction Unit 136 Output Control Unit N Network

Claims (8)

A reception unit that receives input of a first keyword to be processed and input of information specifying a range to be included in the second keyword group to be processed together with the first keyword;
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, A setting unit that extracts a keyword group up to a hierarchy specified by information specifying a range and sets the second keyword group;
Extraction for extracting content information whose degree of match with the first keyword and the second keyword group is greater than or equal to the second storage unit that stores information on the content as summary information for each content And
An output control unit for outputting a list of the extracted content information;
An information output system comprising: The content is a book,
The extraction unit refers to the second storage unit that stores information on the book as bibliographic information for each of the books, and the degree of match with respect to the first keyword and the second keyword group is equal to or greater than a predetermined value. Extract book information,
The information output system according to claim 1. Further, when receiving the input of the first keyword, a third storage unit that stores a value indicating the degree of a plurality of predetermined items for each keyword is referred to and a plurality of keywords including the first keyword are stored. A calculation unit that reads the values stored in association with each other, and calculates a value indicating the closeness of the relationship between the keywords including the first keyword based on the read values;
A generating unit that generates a hierarchical relationship between the first keyword and another keyword having a relationship based on the calculated value indicating the proximity of the relationship;
Have
The output control unit outputs the generated hierarchical relationship to the first storage unit as the related word data.
The information output system according to claim 1 or 2, wherein The first keyword is a first keyword group including a plurality of keywords,
The extraction unit switches a method of extracting information on content having a degree of match of a predetermined value or more according to the number of keywords indicated in the first keyword group and the second keyword group.
The information output system according to any one of claims 1 to 3. The extraction unit extracts information on content having a predetermined degree of match or more in each of a plurality of types of extraction methods,
The output control unit outputs a result of extraction by each of the plurality of types of extraction methods;
The information output system according to any one of claims 1 to 4, wherein: The extraction unit has a score obtained by adding the priority for each content calculated based on one or more of sales importance, author popularity importance, genre, and date to the degree of match. Extract content information,
The information output system according to any one of claims 2 to 5, wherein: Receiving an input of a first keyword to be processed and an input of information for specifying a range to be included in the second keyword group to be processed together with the first keyword;
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, and Extracting a keyword group up to the hierarchy specified by the information specifying the range and setting it as the second keyword group;
For each content, refer to a second storage unit that stores information related to the content as summary information, and extract information on content that has a degree of match with the first keyword and the second keyword group of a predetermined value or more,
Outputting a list of information of the extracted content;
An information output method, wherein a computer executes a process. Receiving an input of a first keyword to be processed and an input of information for specifying a range to be included in the second keyword group to be processed together with the first keyword;
With reference to a first storage unit that stores related word data indicating a hierarchical relationship between keywords in association with a value indicating the closeness of a relationship between two keywords, the first keyword as a starting point, and Extracting a keyword group up to the hierarchy specified by the information specifying the range and setting it as the second keyword group;
For each content, refer to a second storage unit that stores information related to the content as summary information, and extract information on content that has a degree of match with the first keyword and the second keyword group of a predetermined value or more,
Outputting a list of information of the extracted content;
An information output program for causing a computer to execute processing.

Images