JP2013210787A - Book recommendation system, book recommendation method, book recommendation server, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a book recommendation system capable of recommending a book similar to information (in-site) in which a user is really interested.SOLUTION: A book recommendation server 2 retrieves MARC data 25 with a book ID of a selected book and acquires a title 78 (S22). Next, the book recommendation server 2 retrieves a table of in-site data 26 with the acquired title 78 and a user ID and acquires an in-site word 84 having large weight 85 in the retrieved table (S23). Then, the book recommendation server 2 retrieves the MARC data 25 with the acquired in-site word 84 and acquires a book ID 71 as a book ID of a recommended book, that is associated on the title 78 including the same word as the in-site word 84 (S24). After that, the book recommendation server 2 retrieves book data 22 and the MARC data 25 with the book IDs of the selected book and the recommended book and acquires book detail information on the selected book and the recommended book (S25).

Description

  The present invention relates to a book recommendation system that recommends an appropriate book for each user.

  Currently, in a search service such as Google (registered trademark) or an EC (Electronic Commerce) site such as Amazon (registered trademark), a user inputs a “search word” into a predetermined box to obtain desired information and products. You can search. Further, Amazon (registered trademark) has a function of recommending another product to the user A when the user B who has purchased the same product as the product selected by the user A has also purchased another product. However, this other product is a product that user B is interested in and purchased, and is not necessarily a product that user A is interested in.

  In addition, for example, Patent Documents 1 and 2 also disclose a mechanism for recommending information. Patent Document 1 discloses a mechanism for storing purchase histories and purchase cycles of other users and recommending products and the like that match the current user from this data. Moreover, in patent document 2, the item purchase history and item attribute which other users purchased are memorized, the distribution is created, and when a user purchases a certain product, from the distribution, it is surprisingly surprising. A mechanism for recommending products that feel the same is disclosed.

JP 2004-220152 A JP 2011-96025 A

  However, in the mechanism described in Patent Document 1, as with Amazon (registered trademark) and the like, it is recommended based on the preference and purchase cycle of others, so it is not always possible to recommend products that the user is really interested in Not exclusively. Similarly, in the mechanism described in Patent Document 2, since data of another user is used, the user himself / herself cannot always recommend a product that feels surprising in a good sense.

  Hereinafter, information that the user himself / herself is really interested in, including potential awareness, will be referred to as “insight”. Insight is information that is inferred including not only the meaning of the words in the search word but also the user's behavior.

  The present invention has been made in view of the above-described problems, and its purpose is to provide a book recommendation system capable of recommending a book close to information (insight) that the user is really interested in. Etc. is to provide.

  A first invention for achieving the above-described object is a book recommendation system in which a terminal and a book recommendation server for recommending a book are connected via a network, and the book recommendation server includes: Machine-readable catalog data storage means for storing machine-readable catalog data associated with a subject including a word indicating the content of the book is associated with one or more tables for each user, and for each table, the user Search the machine-readable catalog data with the insight data storage means for storing the insight data in which the insight word indicating the information of interest is ranked with weight and the selected book received from the terminal Acquired by the selected book subject acquiring means for acquiring the subject associated with the selected book and the selected book subject acquiring means. Searching the table of the insight data with the subject name, the insight word acquiring means for acquiring the insight word having the large weight in the searched table, and the insight word acquiring means Search the machine-readable catalog data with an insight word, and recommend a book associated with the subject that contains the same word as the insight word, or a book with a title that includes the insight word A book recommendation system comprising: a recommended book acquisition unit that acquires as a book; and a recommended book provision unit that provides the terminal with the recommended book acquired by the recommended book acquisition unit. According to the book recommendation system of the first invention, it is possible to recommend a book close to information that the user is really interested in.

  When the book recommendation server in the first invention further receives a purchase instruction of a book or an instruction to add a desired book from the terminal, the book recommendation server searches the machine-readable catalog data for the purchased book or the desired book, and It is desirable to include an insight data update unit that acquires the subject associated with a desired book and updates the insight data according to a predetermined update rule based on the acquired subject. Accordingly, it is possible to set a large weight for an insight word indicating information that the user is really interested in considering both the user's explicit consciousness and potential consciousness. As a result, it is possible to recommend a book close to information that the user is really interested in.

  The book recommendation server according to the first aspect of the invention further comprises insight data adding means for adding the search word to the insight data according to a predetermined addition rule when a search word is received from the terminal. It is desirable to do. This makes it possible to store insight words in separate tables in consideration of user behavior.

  In the machine-readable catalog data according to the first aspect of the present invention, a book genre and / or publication form is associated with each book, and the recommended book providing means further includes the same genre as the recommended book. It is desirable to provide related books based on the publication format. This makes it possible to provide a book close to the user's insight next to the recommended book among books of the same genre, although the user does not intend. Further, when the publication form is “series”, the recommended book and the related book can be provided to the user in the whole series.

  According to a second aspect of the present invention, there is provided a book recommendation server including a terminal and machine-readable catalog data storage means for storing machine-readable catalog data in which a subject indicating the content of a book is associated with each book, and recommending the book. A book recommendation method in a book recommendation system connected via a network, wherein the book recommendation server is associated with one or more tables for each user, and the user is really interested in each table. An insight data indicating information is stored in the insight data storage step for storing the insight data ranked with weight, and the selected book received from the terminal is searched for the machine-readable inventory data, Acquired by the selected book subject obtaining step for obtaining the associated subject and the selected book subject obtaining step. Searching the table of the insight data by the subject, and acquiring the insight word having the large weight in the searched table, and the insight word acquiring step Search the machine-readable catalog data with an insight word, and recommend a book associated with the subject that contains the same word as the insight word, or a book with a title that includes the insight word A book recommendation method comprising: performing a recommended book acquisition step acquired as a book, and a recommended book providing step of providing the terminal with the recommended book acquired in the recommended book acquisition step. According to the book recommendation method of the second invention, it is possible to recommend a book close to information that the user is really interested in.

  A third invention is a book recommendation server that is connected to a terminal via a network and recommends books, and stores machine-readable catalog data in which a subject indicating the contents of a book is associated with each book. The readable catalog data storage means is associated with one or more tables for each user, and the insight words indicating the information that the user is really interested in are ranked with weights for each table. An insight data storage means for storing insight data; a selected book subject acquisition means for searching the machine-readable catalog data with a selected book received from the terminal and acquiring the subject associated with the selected book; Search the table of the insight data by the subject acquired by the selected book subject acquisition means, the table to be searched The insight word acquisition means for acquiring the insight word having a large weight in the search, and the machine-readable catalog data is searched with the insight word acquired by the insight word acquisition means, the same as the insight word The book with the subject including the word or the book with the book name including the insight word is acquired by the recommended book acquisition unit and the recommended book acquisition unit. A book recommendation server comprising: recommended book provision means for providing the recommended book to the terminal. By the book recommendation server of the third invention, it is possible to construct the book recommendation system of the first invention and execute the book recommendation method of the second invention.

  A fourth aspect of the invention is a program for connecting a computer to a terminal via a network and causing the computer to function as a book recommendation server for recommending a book, wherein the computer has a subject indicating the content of the book for each book. Machine-readable inventory data storage means for storing the associated machine-readable inventory data is associated with one or more tables for each user, and each table indicates information that the user is really interested in. Site data is stored in the insight data storage means for storing the insight data ranked with weight, and the machine-readable catalog data is searched with the selected book received from the terminal, and is associated with the selected book. The selected book subject obtaining means for obtaining the subject and the subject obtained by the selected book subject obtaining means Searching the table of the insight data, the insight word acquisition means for acquiring the insight word having a large weight in the searched table, and the insight word acquired by the insight word acquisition means The machine-readable catalog data is searched, and a book associated with the subject including the same word as the insight word or a book with a book name including the insight word is acquired as a recommended book. It is a program for functioning as a recommended book acquisition unit and a recommended book providing unit that provides the terminal with the recommended book acquired by the recommended book acquisition unit. The recommended book server of the third invention can be obtained by installing the program of the fourth invention on a general-purpose computer.

  According to the present invention, it is possible to provide a book recommendation system or the like that can recommend a book close to information (insight) that a user is really interested in.

Configuration diagram of book recommendation system 1 Computer hardware configuration diagram The figure which shows an example of the data memorize | stored in the database 3 The figure which shows an example of the user data 21 The figure which shows an example of the book data 22 The figure which shows an example of the desired book data 23 The figure which shows an example of the purchase history data 24 The figure which shows an example of MARC data 25 The figure which shows an example of the insight data 26 Flowchart showing the process flow of terminal 4 Flow chart showing the flow of search processing Flow chart showing the flow of product detail display processing Flow chart showing the flow of purchase processing Flow chart showing the flow of desired book addition process Diagram showing insight data 26a Diagram showing insight data 26b Diagram showing insight data 26c Diagram showing insight data 26d

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The book targeted by the book recommendation system of the present invention may be an electronic book or a paper book. The book recommendation system of the present invention may be applied to a book sales system or a simple book search system. Hereinafter, an online electronic book sales system will be described as an example. In the following, the term “book” may refer to a book body or information that identifies a book.

  First, the configuration and operation of the book recommendation system 1 will be described in detail with reference to FIGS.

  FIG. 1 is a configuration diagram of the book recommendation system 1. As shown in FIG. 1, the book recommendation system 1 is configured by connecting a book recommendation server 2 and a terminal 4 via a network 5.

  The book recommendation system 1 is a server that recommends an optimal book for each user who uses the terminal 4. The book recommendation server 2 has a database 3. The database 3 stores at least machine readable catalog (MARC) data and insight data.

Machine-readable catalog (hereinafter referred to as “MARC”) data is a medium that can be processed by a computer in a certain format for information described in the catalog entry such as bibliographic descriptions, headings, and location symbols of books. Is recorded. As an example, “TRC” of Library Information Distribution Center, Inc.
MARC "etc. are mentioned. In the MARC data, at least a subject indicating the contents of the book is associated with each book. The subject includes one or more words. For example, the title for a book “If a high school baseball girl manager reads Drucker's“ Management ”” (Natsuki Iwasaki, Diamond, December 2009) includes words such as “Management” and “Coaching”. .

  Insight data is data for knowing information that is really interested, including potential awareness. The insight data is associated with one or more tables for each user. Further, in the insight data, for each table, insight words indicating information that the user is really interested in are ranked with weights. The weight is a selection criterion for the insight word used in the recommended book acquisition process.

  Insight words are added in response to user actions. The user's behavior is, for example, “input a search word to search for a book”, “select one book and browse detailed information of the book”, and the like. As the insight word, a search word input by the user at the terminal 4, a word included in the subject of the MARC data, or the like is registered.

  The book recommendation server 2 searches the machine-readable catalog data with the selected book received from the terminal 4 and acquires the subject associated with the selected book. The selected book is a book that the user himself selects on the screen displayed on the terminal 4.

  Next, the book recommendation server 2 searches the table of the insight data with the acquired subject, acquires the insight word having a large weight in the searched table, and stores the machine-readable catalog data with the acquired insight word. The search is performed, a book associated with the same subject as the insight word, or a book with a book name including the insight word is acquired as a recommended book, and the recommended book is transmitted to the terminal 4. The recommended book is a book that is recommended by the book recommendation server 2 and is close to information (insight) that the user is really interested in.

  The terminal 4 is a PC (Personal Computer), a mobile phone, a smartphone, a tablet terminal, etc. owned by the user. The terminal 4 transmits data input by the user to the book recommendation server 2 and displays data transmitted from the book recommendation server 2.

  FIG. 2 is a hardware configuration diagram of the computer. Note that the hardware configuration in FIG. 2 is an example, and various configurations can be adopted depending on the application and purpose.

  A computer that implements the book recommendation server 2, the terminal 4, and the like is connected to the control unit 11, the storage unit 12, the input unit 13, the display unit 14, the communication control unit 15, and the like via the bus 16.

  The control unit 11 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The CPU calls a program stored in the storage unit 12, ROM, recording medium, etc. to a work memory area on the RAM, executes it, drives and controls each device connected via the bus 16, and is described later by the computer. Realize processing. The ROM is a non-volatile memory and permanently holds a computer boot program, a program such as BIOS, data, and the like. The RAM is a volatile memory, and temporarily stores programs, data, and the like loaded from the storage unit 12, ROM, recording medium, and the like, and includes a work area used by the control unit 11 for performing various processes.

  The storage unit 12 is, for example, an HDD (Hard Disk Drive), a flash memory, or the like, and stores a program executed by the control unit 11, data necessary for program execution, an OS (Operating System), and the like. As for the program, a control program corresponding to the OS and an application program for causing a computer to execute processing to be described later are stored. Each of these program codes is read by the control unit 11 as necessary, transferred to the RAM, read by the CPU, and executed as various means.

  The input unit 13 inputs data, and includes input devices such as a keyboard, a pointing device such as a mouse and a touch pen, a numeric keypad, a button, and a switch. The user can perform operation instructions, operation instructions, data input, and the like on the computer via the input unit 13. The display unit 14 includes a display device such as a liquid crystal panel or an organic EL, and a logic circuit or the like (video adapter or the like) for realizing the video function of the computer in cooperation with the display device. The input unit 13 and the display unit 14 may be integrated like a touch panel display.

  The communication control unit 15 includes a communication control device, a communication port, and the like, is a communication interface that mediates communication between the computer and the network 5, and performs communication control between other computers via the network 5. The network 5 may be wired or wireless. The bus 16 is a path that mediates transmission / reception of control signals, data signals, and the like between the devices.

  Hereinafter, details of data used in the book recommendation system 1 will be described with reference to FIGS.

  FIG. 3 is a diagram illustrating an example of data stored in the database 3. The database 3 stores user data 21, book data 22, desired book data 23, purchase history data 24, MARC data 25, insight data 26, and the like.

  FIG. 4 is a diagram illustrating an example of the user data 21. The user data 21 is information of a user who is registered in the electronic book sales system. The user data 21 has data items such as a user name 32 with the user ID 31 as a main key. The user ID 31 is a number or a character string that uniquely identifies the user. The user name is the user's name or nickname. Note that the user data 21 may include information such as the user's age, sex, and address.

  FIG. 5 is a diagram illustrating an example of the book data 22. The book data 22 is data of a book handled in the electronic book sales system. The book data 22 has a book ID 41 as a main key and data items such as a book name 42 and a price 43. The book ID 41 is a number or a character string that uniquely identifies the book. The book name 42 is the title of the book. The price 43 is the price for one book. The book data 22 may include information such as a content introduction sentence, a publisher, a publication date, a genre, and a publication form, as in the MARC data 25 described later.

  FIG. 6 is a diagram illustrating an example of desired book data 23. The desired book data 23 is a list of books that the user wants. The desired book data 23 has data items such as an addition date and time 53 with the user ID 51 and the book ID 52 as main keys. The user ID 51 is the same as the user ID 31. The book ID 52 is the same as the book ID 42. The addition date and time 53 is the date and time when the user has given an instruction to add a book. The desired book data 23 may include information such as user comments.

  FIG. 7 is a diagram illustrating an example of the purchase history data 24. The purchase history data 24 is a history of books purchased by the user. The purchase history data 24 has data items such as purchase date and time with the user ID 61 and the book ID 62 as main keys. The user ID 61 is the same as the user ID 31. The book ID 62 is the same as the book ID 42. The purchase date and time 63 is the date and time when the user gives an instruction to purchase a book. The purchase history data 24 may include information such as user comments and purchase prices.

  FIG. 8 is a diagram illustrating an example of the MARC data 25. As described above, the MARC data 25 is information that is described in the bibliographic description of the book, headings, location symbols, and the like. The MARC data 25 has data items such as a book name 72, a content introduction 73, a publisher 74, a publication date 75, a genre 76, a publication form 77, a subject 78, etc., with the book ID 71 as a main key. The book ID 71 is the same as the book ID 41. The book name 72 is the same as the book name 42.

  The content introduction 73 is an introduction sentence for introducing the content of the book. Publisher 74 is the name of the company that published the book. The publication date 75 is the date when the book was published.

  The genre 76 is a classification according to the contents of the book. Examples of the genre 76 include “literary / non-fiction”, “hobby / life”, “business / education / society”, and the like. The publication form 77 is information indicating in what form the book is published. Examples of the publication form 77 include “single item”, “series”, “reprint”, and the like. As described above, the subject 78 is information indicating the contents of the book. The subject 78 includes one or more words.

  In the above description, the book data 22 and the MARC data 25 are separate data, but both may be integrated. For example, the integrated data of the book data 22 and the MARC data 25 has data items such as a book ID, a price, a content introduction, a publisher, a publication date, a genre, a publication form, and a subject, with the book ID as a main key. Hereinafter, in order to avoid confusion, the book data 22 and the MARC data 25 will be described as separate data.

  FIG. 9 is a diagram illustrating an example of the insight data 26. As described above, the insight data 26 is data for knowing information that is really interested including potential awareness. The insight data 26 has data items such as an insight word 84, a weight 85, a search addition flag 86, and a subject addition flag 87, with the user ID 81, the table ID 82, and the rank 83 as main keys. The user ID 81 is the same as the user ID 31.

  As described above, the insight data 26 is associated with one or more tables for each user. The table ID 82 is a number or a character string that uniquely identifies the table.

  As described above, the insight word 84 is a word indicating information that the user is really interested in. Note that the same insight word 84 may be stored in different tables even for the same user. For example, if you search for “Tohoku” (for example, AND search for “Tohoku” and “Ski resort”) as a word related to hobbies, and AND search for “Tohoku” (for example, “Tohoku” and “Company”) for words related to business ), The books that match the information the user is interested in are different, so they are stored in separate tables. For example, the book recommendation server 2 considers search words input together, the genre of the book, other insight words registered in the table, and stores them in a table having a closer meaning.

  As described above, in the insight data 26, the insight word 84 is ranked for each table. The rank 83 is a number indicating the rank.

  As described above, in the insight data 26, the insight word 84 has a weight. The weight 85 is a numerical value indicating the weight. As described above, the weight 85 is a selection criterion for the insight word 84 used in the recommended book acquisition process.

  The rank 83 is the order in which the weight 85 is large. Therefore, when the value of the weight 85 changes, the rank 83 also changes. Therefore, in order to simplify the data update process, instead of the rank 83, a record ID indicating a number for uniquely identifying a record in the table may be used. The record ID does not change even if the value of the weight 85 changes.

  As described above, the insight word 84 is added in response to a user action. The user's behavior is, for example, “input a search word to search for a book”, “select one book and browse detailed information of the book”, and the like. For example, as the insight word 84, a search word input by the user on the terminal 4 or a word included in the subject 78 of the MARC data 25 related to the book specified by the operation of the terminal 4 by the user is registered. . The search addition flag 86 is a flag indicating whether or not the insight word 84 has been added as a search word. The subject addition flag 87 is a flag indicating whether or not the insight word 84 is added as a word included in the subject 78.

  Below, the detail of operation | movement in the book recommendation system 1 is demonstrated, referring FIGS. 10-14.

  FIG. 10 is a flowchart showing a process flow of the terminal 4. FIG. 10 illustrates a flow in which the terminal 4 receives an input event via the input unit 13 and executes processing according to the received input event. In FIG. 10, only main input events are shown.

  The electronic book sales system sells electronic books on a website, for example. The terminal 4 executes processing to be described later while logged in to a website provided by the electronic book sales system.

  As shown in FIG. 10, when receiving an input event (S1), the control unit 11 of the terminal 4 determines the input event (S2). Note that the processing of S1 and S2 is realized by a function provided by the OS or web browsing software.

  When the input event is “search”, the control unit 11 of the terminal 4 executes a search process in cooperation with the book recommendation server 2 (S3). When the search process ends, the control unit 11 of the terminal 4 transitions to an input event reception state.

  FIG. 11 is a flowchart showing the flow of search processing. As illustrated in FIG. 11, when receiving the search word via the input unit 13, the control unit 11 of the terminal 4 transmits the search word to the book recommendation server 2 (S 11).

  When receiving the search word, the control unit 11 of the book recommendation server 2 searches the book data 22 and the MARC data 25 with the search word, and acquires book summary information (S12). For example, the control unit 11 of the book recommendation server 2 searches the book name 42 (book name 72), the content introduction 73, the subject name 78, etc. using the search word, and the book name 42 (book name 72) and price 43 of the matching data. , Genre 76 and the like are acquired as book summary information.

  Next, the control unit 11 of the book recommendation server 2 adds a search word to the insight data 26 according to a predetermined addition rule (S13).

  An example of the rules for adding insight data 26 will be described. If there is no insight data 26 table associated with the target user, the control unit 11 of the book recommendation server 2 adds a new table. The control unit 11 of the book recommendation server 2 sets the insight word 84 = search word, weight 85 = predetermined value by search addition, search addition flag 86 = Yes, and subject addition flag 87 = No for the new table. Add only one record.

  In the addition rule described above, whether or not the search word exists in the existing table is used as a determination criterion. However, the insight data 26 may be added for each genre 76 of the book. In this case, for example, in the insight data 26, an item corresponding to the genre 76 is set as one of the main keys.

  Next, the control unit 11 of the book recommendation server 2 transmits a list of book summary information to the terminal 4 (S14). The control unit 11 of the terminal 4 displays a list of book summary information on the display unit 14 (S15).

  Returning to the description of FIG. When the input event is “product selection”, the control unit 11 of the terminal 4 performs product detail display processing in cooperation with the book recommendation server 2 (S4). When the product detail display process ends, the control unit 11 of the terminal 4 transitions to an input event reception state.

  FIG. 12 is a flowchart showing the flow of the product detail display process. As shown in FIG. 12, when the control unit 11 of the terminal 4 receives the selected book via the input unit 13, the control unit 11 transmits the book ID of the selected book to the book recommendation server 2 (S21).

  When receiving the book ID of the selected book, the control unit 11 of the book recommendation server 2 searches the MARC data 25 with the book ID of the selected book and acquires the subject 78 (S22). Next, the control unit 11 of the book recommendation server 2 searches the table of the insight data 26 using the acquired subject 78 and user ID, and acquires the insight word 84 having a large weight 85 in the searched table (S23). ). Next, the control unit 11 of the book recommendation server 2 searches the MARC data 25 with the acquired insight word 84, and the book ID 71 of the book associated with the subject 78 including the same word as the insight word 84. Alternatively, the book ID 71 of the book with the book name 72 including the insight word 84 is acquired as the book ID of the recommended book (S24). The number of recommended books is one or more.

  The recommended ranking of the recommended books is, for example, in the order closer to the user's insight in consideration of the weight 85 related to the insight word 84 used in the recommended book acquisition process, the search addition flag 86, the subject addition flag 87, and the like. For example, if the weight 85 is large, the control unit 11 of the book recommendation server 2 determines that it is close to the user's insight. This is because the weight 85 is updated so as to become a larger value in the order closer to the user's insight, due to the user's behavior, by a process described later. Further, the control unit 11 of the book recommendation server 2 determines that both the search addition flag 86 and the subject addition flag 87 are closer to the user's insight even if the weight 85 is the same. This is because the obvious consciousness of the user can be taken into account by the search word, and the latent consciousness of the user can be taken into consideration by the subject line 78 showing the contents of the book accurately.

  In the above description, S22 to S24 have been described as separate processes. However, these processes may be realized by one SQL statement. SQL is a database language in a relational database management system.

  Next, the control unit 11 of the book recommendation server 2 searches the book data 22 and the MARC data 25 with the book ID of the selected book and the recommended book, and acquires the book detailed information of the selected book and the recommended book (S25). The detailed book information includes, for example, a book name 42 (book name 72), a price 43, a content introduction 73, a publisher 74, a publication date 75, a genre 76, and a publication form 77.

  Next, the control unit 11 of the book recommendation server 2 transmits the selected book and the book detailed information of the recommended book to the terminal 4 (S26). The control unit 11 of the terminal 4 displays the book detail information of the selected book and the recommended book on the display unit 14 (S27).

  Here, the control unit 11 of the book recommendation server 2 searches for a related book that is the same genre 76 as the recommended book based on the genre 76 of the recommended book and is close to the user's insight, and transmits it to the terminal 4. May be. Thereby, although the user does not intend, among books of the same genre, a book close to the user's insight can be provided next to the recommended book.

  Further, the control unit 11 of the book recommendation server 2 refers to the recommended book publication form 77, and when the publication form 77 = “series”, the book of the same series is transmitted to the terminal 4 as a related book. good. Thus, the recommended book and the related book can be combined and provided to the user in the entire series.

  Returning to the description of FIG. When the input event is “purchase”, the control unit 11 of the terminal 4 executes a purchase process in cooperation with the book recommendation server 2 (S5). When the purchase process ends, the control unit 11 of the terminal 4 transitions to an input event reception state.

  FIG. 13 is a flowchart showing the flow of purchase processing. As illustrated in FIG. 13, when receiving the purchased book via the input unit 13, the control unit 11 of the terminal 4 transmits the book ID of the purchased book to the book recommendation server 2 (S 31).

  When receiving the book ID of the purchased book, the control unit 11 of the book recommendation server 2 executes a settlement process related to the purchased book (S32). Since the payment process may use a known technique, the description thereof is omitted. When the settlement process ends, the control unit 11 of the book recommendation server 2 updates the purchase history data 24 (S33).

  Next, the control unit 11 of the book recommendation server 2 searches the MARC data 25 by the book ID of the purchased book, acquires the subject 78 (S34), searches the insight data 26 by the acquired subject 78, and determines the predetermined number. The insight data 26 is updated in accordance with the update rule (S35).

  An example of the update rule for the insight data 26 will be described. The update of the insight data 26 means both adding a new record to the table of the insight data 26 and updating the weight 85 of the existing record. That is, the control unit 11 of the book recommendation server 2 does not include the same insight word 84 as the word included in the subject 78 acquired in S34 in the table of the insight data 26 associated with the target user. Add a new record. Further, the control unit 11 of the book recommendation server 2 includes the same insight word 84 as the word included in the subject 78 acquired in S34 in the table of the insight data 26 associated with the target user. For example, the weight 85 of the record is updated.

  The update rule has, for example, the following three patterns. Note that the control unit 11 of the book recommendation server 2 sets the “first value” that is the largest predetermined value, the “second value” that is the next largest predetermined value, and the first for every three patterns. One of the “third values” which is a small predetermined value is added to the weight 85.

  In the first pattern, if the word included in the subject 78 acquired in S34 is the same as the search word (search addition flag 86 = Yes word), the control unit 11 of the book recommendation server 2 uses the same word. In contrast, the “first value” is added to the weight 85.

  In the second pattern, if the word included in the subject 78 acquired in S34 is the same as the search word (search addition flag 86 = Yes word), the control unit 11 of the book recommendation server 2 uses the same word. The “second value” is added to the weight 85 for other words that are not.

  In the third pattern, the control unit 11 of the book recommendation server 2 determines that the word included in the subject 78 acquired in S34 is not the same as the search word (search addition flag 86 = Yes word). The “third value” is added to the weight 85 for the word.

  By updating the weight 85 in accordance with the above update rule, the insight word 84 indicating the information (insight) that the user is really interested in is considered in consideration of both the explicit consciousness and the potential consciousness of the user. On the other hand, a large weight 85 can be set. As a result, it is possible to recommend a book close to information that the user is really interested in.

  Next, the control unit 11 of the book recommendation server 2 transmits a purchase completion notification to the terminal 4 (S36). The control unit 11 of the terminal 4 displays a purchase completion notification on the display unit 14 (S37).

  Returning to the description of FIG. When the input event is “Add desired book”, the control unit 11 of the terminal 4 executes a desired book addition process in cooperation with the book recommendation server 2 (S6). When the desired book addition process is completed, the control unit 11 of the terminal 4 transitions to an input event reception state.

  FIG. 14 is a flowchart showing the flow of desired book addition processing. As illustrated in FIG. 14, when the control unit 11 of the terminal 4 receives a desired book via the input unit 13, the control unit 11 transmits the book ID of the desired book to the book recommendation server 2 (S 41).

  When receiving the book ID of the desired book, the control unit 11 of the book recommendation server 2 updates the desired book data 23 (S42).

  Next, the control unit 11 of the book recommendation server 2 searches the MARC data 25 by the book ID of the book desired, acquires the subject 78 (S43), searches the insight data 26 by the acquired subject 78, and performs predetermined processing. The insight data 26 is updated according to the update rule (S44). The update rule of the insight data 26 is the same as the update process (S35) based on the purchased book shown in FIG.

  Next, the control unit 11 of the book recommendation server 2 transmits a desired book addition completion notification to the terminal 4 (S45). The control unit 11 of the terminal 4 displays a desired book addition completion notification on the display unit 14 (S46).

  Returning to the description of FIG. When the input event is “end”, the control unit 11 of the terminal 4 executes an end process and logs out from the website provided by the electronic book sales system (S7).

  In the above description, the addition process (S13) and the update process (S34 and S35, and S43 and S44) of the insight data 26 are executed in real time format. However, these processes may be executed in batch format. . The batch format is a format that is scheduled and executed every predetermined period (for example, an hour unit, a day unit, etc.).

  For example, for the process of adding the insight data 26, the book recommendation server 2 stores search history data (not shown) for each user, extracts one search word from the search history data, and corresponds to S13. Execute the process.

  Further, for example, for the update process of the insight data 26, the book recommendation server 2 extracts the book ID 62 one by one from the purchase history data 24 stored for each user, and executes a process corresponding to S34 and S35. . Further, the book recommendation server 2 extracts book IDs 52 one by one from the book data 23 desired to be stored for each user, and executes processing corresponding to S43 and S44.

  Hereinafter, specific examples of the insight data addition processing and update processing will be described with reference to FIGS. 15 to 18.

  FIG. 15 shows the insight data 26a. In the insight data 26a shown in FIG. 15, in order to search for books related to business, a user with a user ID 81 = “1” performs a search by inputting a search word = “logical thinking”, and then some books Shows the result of adding or purchasing as a desired book. In the example shown in FIG. 15, the record with rank 83 = "1" has an insight word 84 = "logical thinking", a weight 85 = "1.38", a search addition flag 86 = "Yes", and a subject addition flag 87 = “Yes”. The reason why both the search addition flag 86 and the subject addition flag 87 are “Yes” is because “subject” is included in the subject 78 related to the desired book or purchased book. Further, the record of rank 83 = 2 is “insight word 84 =“ presentation power ”, weight 85 =“ 1.21 ”, search addition flag 86 =“ No ”, and subject addition flag 87 =“ Yes ”. ing.

  FIG. 16 shows the insight data 26b. In the insight data 26b shown in FIG. 16, in order to search for books related to hobbies, the user with the user ID 81 = “1” performs a search by inputting the search word = “ski”, and then some books are searched. It shows the result of adding or purchasing as a book you want. In this case, a table with table ID 82 = “2” shown in FIG. 16 is added as a table different from the table with table ID 82 = “1” shown in FIG. In the example shown in FIG. 16, the record with ranking 83 = “1” has an insight word 84 of “ski”, a weight of 85 = 1.38, a search addition flag 86 = “Yes”, and a subject addition flag 87 = “ “Yes”. Further, the record of rank 83 = “2” has the insight word 84 = “Naeba”, the weight 85 = “1.21”, the search addition flag 86 = “No”, and the subject addition flag 87 = “Yes”. Yes.

  FIG. 17 is a diagram showing the insight data 26c. The insight data 26c shown in FIG. 17 shows the current state of the table associated with the user with the user ID 81 = “2”. In the example shown in FIG. 17, there are three records of the insight word 84 = “nurture subordinates”, “management”, and “coaching”.

  In the state of FIG. 17, when a user with a user ID 81 = “2” is looking for a reference book for instructing a company subordinate, “If a high school baseball girl manager reads Drucker's“ Management ” ”(Natsumi Iwasaki, Diamond, December 2009) (hereinafter referred to as“ Moshidora ”) is selected. Then, the book recommendation server 2 acquires “management” and “coaching”, which are the subject 78 of “if Dora”. Next, the book recommendation server 2 searches the insight data 26 by “management” and “coaching”. As a result of this search, a table of insight data 26c shown in FIG. 17 is obtained. Next, the book recommendation server 2 acquires “Grow subordinates”, “Management”, and “Coaching”, which are the insight words 84 having a large weight in the table of the insight data 26c shown in FIG. Then, the book recommendation server 2 searches the MARC data 25 for “grow subordinates”, “management”, and “coaching”, and the book with the same subject 78 or the book name including the insight word. As a recommended book.

  As an example of a book acquired as a recommended book, for example, a book “Coaching-Magic of Words and Beliefs” (Hiromitsu Ochiai, Diamond Corporation, September 2001) can be cited. Since the user has been searching for a book that can be used as a reference for instructing the subordinates of the company, this book can be said to be a book close to information that the user is really interested in.

  Further, it is assumed that the user with the user ID 81 = “2” has purchased a book “Coaching—a magic of words and beliefs”. Then, the book recommendation server 2 acquires “coaching” and “baseball”, which are subject lines 78 of “coaching—magic of words and beliefs”. Next, the book recommendation server 2 searches the insight data 26 with “coaching” and “baseball”. As a result of this search, a table of insight data 26c shown in FIG. 17 is obtained. Then, the book recommendation server 2 updates the insight data according to the update rule.

  FIG. 18 is a diagram showing the insight data 26d. The insight data 26d shown in FIG. 18 shows the update result resulting from the purchase of the book “Coaching-Magic of Words and Beliefs”. In the insight data 26d, the weight 85 related to the insight word 84 = “coaching” is changed from 1.20 to 1.40. In the insight data 26d, a record of insight word 84 = “baseball” is added.

  As described above, in the book recommendation system 1 according to the embodiment of the present invention, it is possible to recommend a book close to information that the user is really interested in.

  The preferred embodiments of the book recommendation system and the like according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these naturally belong to the technical scope of the present invention. Understood.

1 ……… Book recommendation system 2 ……… Book recommendation server 3 ……… Database 4 ……… Terminal 5 ……… Network 21 ……… User data 22 ……… Book data 23 ……… Book data 24… …… Purchase history data 25 ……… MARC data 26 ……… Insight data

Claims (7)

A book recommendation system in which a terminal and a book recommendation server for recommending books are connected via a network,
The book recommendation server is
Machine-readable catalog data storage means for storing machine-readable catalog data associated with a subject including a word indicating the content of the book for each book;
Each user is associated with one or more tables, and each table stores insight data in which insight words indicating information that the user is really interested in are ranked with weights. Site data storage means;
Searching the machine-readable catalog data with a selected book received from the terminal, and a selected book subject acquisition means for acquiring the subject associated with the selected book;
Searching for the table of the insight data by the subject acquired by the selected book subject acquiring means, and acquiring the insight word having a large weight in the searched table;
The machine-readable catalog data is searched with the insight word acquired by the insight word acquisition means, and the book associated with the subject including the same word as the insight word, or the insight A recommended book acquisition means for acquiring a book with a title including a site word as a recommended book;
Recommended book providing means for providing the terminal with the recommended book acquired by the recommended book acquiring means;
A book recommendation system comprising: The book recommendation server further includes:
Upon receiving a book purchase instruction or a desired book addition instruction from the terminal, the machine-readable catalog data is searched for the purchased book or the desired book, and the subject associated with the purchased book or the desired book is acquired. , Insight data update means for updating the insight data according to a predetermined update rule based on the acquired subject,
The book recommendation system according to claim 1, further comprising: The book recommendation server further includes:
When receiving a search word from the terminal, according to a predetermined addition rule, insight data addition means for adding the search word to the insight data,
The book recommendation system according to claim 1, wherein the book recommendation system is provided. In the machine-readable catalog data, for each book, a book genre and / or publication form is associated,
The book recommendation system according to any one of claims 1 to 3, wherein the recommended book providing means further provides a related book based on the same genre and / or publication form as the recommended book. . A terminal is provided with machine-readable catalog data storage means for storing machine-readable catalog data in which a subject indicating the content of a book is associated with each book, and a book recommendation server that recommends the book is connected via a network. A book recommendation method in a book recommendation system,
The book recommendation server is
Each user is associated with one or more tables, and each table stores insight data in which insight words indicating information that the user is really interested in are ranked with weights. A site data storage step;
Searching the machine-readable catalog data with selected books received from the terminal, and acquiring a selected book subject that acquires the subject associated with the selected book;
Searching the table of the insight data with the subject acquired by the selected book subject acquiring step, and acquiring the insight word with the large weight in the searched table,
The machine-readable catalog data is searched with the insight word acquired by the insight word acquisition step, and the book associated with the subject including the same word as the insight word, or the insight A recommended book acquisition step of acquiring a book with a book name including a site word as a recommended book;
A recommended book providing step of providing the recommended book acquired by the recommended book acquiring step to the terminal;
A method for recommending books. A book recommendation server connected to a terminal via a network and recommending books,
Machine-readable inventory data storage means for storing machine-readable inventory data in which a subject indicating the content of the book is associated with each book;
Each user is associated with one or more tables, and each table stores insight data in which insight words indicating information that the user is really interested in are ranked with weights. Site data storage means;
Searching the machine-readable catalog data with a selected book received from the terminal, and a selected book subject acquisition means for acquiring the subject associated with the selected book;
Searching for the table of the insight data by the subject acquired by the selected book subject acquiring means, and acquiring the insight word having a large weight in the searched table;
The machine-readable catalog data is searched with the insight word acquired by the insight word acquisition means, and the book associated with the subject including the same word as the insight word, or the insight A recommended book acquisition means for acquiring a book with a title including a site word as a recommended book;
Recommended book providing means for providing the terminal with the recommended book acquired by the recommended book acquiring means;
A book recommendation server comprising: A program for connecting a computer to a terminal via a network and causing a computer to function as a book recommendation server for recommending a book,
The computer,
Machine-readable inventory data storage means for storing machine-readable inventory data in which a subject indicating the content of the book is associated with each book;
Each user is associated with one or more tables, and each table stores insight data in which insight words indicating information that the user is really interested in are ranked with weights. Site data storage means;
Searching the machine-readable catalog data with a selected book received from the terminal, and a selected book subject acquisition means for acquiring the subject associated with the selected book;
Searching for the table of the insight data by the subject acquired by the selected book subject acquiring means, and acquiring the insight word having a large weight in the searched table;
The machine-readable catalog data is searched with the insight word acquired by the insight word acquisition means, and the book associated with the subject including the same word as the insight word, or the insight A recommended book acquisition means for acquiring a book with a title including a site word as a recommended book;
Recommended book providing means for providing the recommended book acquired by the recommended book acquiring means to the terminal;
Program to function as.

Images