JP5293970B2 - Product recommendation method and product recommendation system - Google Patents

Description

  The present invention relates to a product recommendation method and a product recommendation system for making recommendations that match consumer trends.

  Conventionally, a technique for classifying consumers based on consumption history information of a plurality of consumers for a plurality of products has been proposed. For example, in the prior art described in Patent Document 1, the server device stores purchase history information as consumption history information indicating a purchase history of a product purchased by a purchaser as a consumer stored in a purchase history information storage unit. Based on this, a product whose number of purchasers is equal to or greater than a preset threshold is identified as a trendy product. Then, purchasers who have purchased the product at an early stage satisfying a certain period of time among the purchasers of the trendy product are classified as trend leaders. Thereby, the product can be recommended from the purchase behavior of the product of the trend leader.

JP 2008-198163 A

  In recent years, it has been considered to make recommendations that match consumer trends for products related to a business using consumption history information of a plurality of consumers for a plurality of products related to a business. At that time, for example, using the consumption history information of a plurality of consumers for a plurality of products accumulated by a technique such as the above-described prior art, it is possible to make a recommendation that matches the consumer trend for the product. At this time, if there is abundant consumption history information about the consumer's products, it is possible to make an effective recommendation that accurately matches the consumer's trend. Conversely, if the consumer history information about the consumer's products is not abundant, there is a problem that a recommendation that does not match the consumer's trend is made.

  An object of the present invention is a product recommendation method capable of making an effective recommendation that accurately matches the consumer's trend for the product even when there is not much consumption history information about the product of the consumer And providing a product recommendation system.

  In order to achieve the above object, the product recommendation method of the first invention associates identification information of a plurality of first products related to the first business with consumption behavior histories of a plurality of consumers for the first product. Accessed to a first consumer database that stores consumption history information for each first product, and acquired in a history acquisition procedure for acquiring consumption history information of a plurality of consumers for a plurality of first products, and acquired in the history acquisition procedure A clustering procedure for generating clustering information in which each consumer's identification information is associated with any one of a plurality of units using the consumption history information of the plurality of consumers for the plurality of first products; Consumption history information in which identification information of a plurality of second products related to business is associated with consumption behavior histories of a plurality of consumers or specific consumption-related behavior histories other than consumption for the second products The second consumer database stored for each second product is accessed, and the identification information of each of the plurality of consumers who has performed a specific consumption-related behavior other than consumption behavior or consumption on the predetermined second product The identification information acquisition procedure to be acquired and the clustering database that stores the clustering information generated in the clustering procedure are accessed, and the number of consumers whose identification information has been acquired in the identification information acquisition procedure is counted for each unit. And at least one consumer information terminal among a plurality of consumers associated with the unit according to the consumer counting procedure and the number of consumers counted for each unit in the consumer counting procedure Recommendation for determining recommendation information related to at least the consumption behavior or the consumption-related behavior A recommended information output procedure for obtaining the recommended information determined in the recommended information determining procedure from the recommended database, and outputting the acquired recommended information to the information terminal of the at least one consumer; It is characterized by having.

  In the first invention of the present application, identification information of a plurality of first products related to the first business is associated with consumption behavior histories of a plurality of consumers for the first product in advance, and the first consumer database is used as consumption history information. Is remembered. In the history acquisition procedure, the first consumer database is accessed to acquire consumption history information of a plurality of consumers for a plurality of first products. Using the acquired consumption history information, clustering information is generated by associating each consumer's identification information with one of a plurality of units in a clustering procedure.

  On the other hand, the first invention of the present application also includes, in advance, identification information of a plurality of second products related to the second business, which is different from the first business, and consumption behavior histories of a plurality of consumers for the second product or a specific The consumption-related behavior history is associated with and stored in the second consumer database as consumption history information. In the identification information acquisition procedure, the second consumer database is accessed to identify a plurality of consumers who have performed consumption actions or consumption related actions on the predetermined second product, and the plurality of consumers' Identification information is acquired. When consumers are identified in this way, the clustering database is accessed in the consumer counting procedure, and the number of consumers whose identification information has been acquired in the identification information acquisition procedure is counted for each unit. As a result, the number of consumers who have performed consumption behavior or consumption-related behavior for the specific second product is counted in each unit generated by clustering related to the first product. Thereafter, in the recommended information determination procedure, recommended information for the information terminal of at least one consumer of the unit is determined according to the count result for each unit. The determined recommended information is acquired from the recommended database in the recommended information output procedure, and is output to the information terminal of the at least one consumer.

  As a result, for each unit based on the clustering for the first product, recommended information is determined according to the number of consumers who have performed consumption behavior on the second product or performed consumption-related behavior. Output to the consumer's information terminal. As a result, even when there is not much consumer consumption history information about the second product, clustering is performed using the first product with abundant consumption history information, and information recommended using the clustering result By determining the contents of, effective recommendations can be made that accurately match the consumer's trend for the second product.

  The product recommendation method according to a second aspect of the present invention is the product recommendation method according to the first aspect, wherein the recommended information determination procedure includes a number of consumers who have performed the consumption behavior or the specific consumption-related behavior for each unit. The recommendation information related to the product is determined so as to be prioritized over the recommended information related to the second product with a small number of consumers who have performed the consumption behavior or the specific consumption-related behavior.

  As a result, for each unit, the recommended information for the second product with a large number of consumers who have performed consumption behavior or performed consumption-related behavior is prioritized over the recommended information for other second products. Is done. As a result, it is possible to make an effective recommendation that reflects the fact that a relatively large number of consumers in the unit are performing consumption behaviors on the second product.

  The product recommendation method according to a third aspect of the present invention is the first or second aspect of the invention, wherein the recommended information output procedure associates the recommended information determined in the recommended information determination procedure for each unit with the unit. It outputs to the said information terminal of the consumer who is not performing the said consumption action or the said specific consumption related action with respect to the said 2nd goods regarding the said recommendation information, It is characterized by the above-mentioned.

  As a result, in each unit, the recommended information about the second product for which a relatively large number of consumers have performed the consumption behavior or the consumption related behavior is obtained from the consumers who have not performed the consumption behavior or the consumption related behavior for the second product. Output to the terminal. As a result, it is possible to make a recommendation targeting consumers who are not performing consumption behavior or consumption-related behavior for the second product at that time. That is, finally, effective recommendation can be reliably performed so that all consumers in one unit having the same preference complete the consumption behavior and the consumption-related behavior for the second product.

  The product recommendation method according to a fourth aspect of the present invention is the product recommendation method according to any one of the first to third aspects, wherein an authentication request reception procedure for receiving an authentication request signal from the information terminal of a specific consumer, Based on the received authentication request signal, an authentication procedure for accessing the authentication database in which the authentication information of the specific consumer is stored in advance with respect to the first business, and performing authentication of the specific consumer related to the authentication request signal And the recommended information output procedure includes the recommended information determined in the recommended information determination procedure for the specific consumer authenticated in the authentication procedure, for the information terminal of the specific consumer The display information is output to the information terminal of the specific consumer.

  In the fourth invention of the present application, authentication information of a specific consumer is stored in the authentication database in advance for the first business. In the authentication request receiving procedure, an authentication request signal from an information terminal of a specific consumer is received. Based on the received authentication request signal, the authentication database is accessed in an authentication procedure, and a specific consumer related to the authentication request signal is authenticated. Then, in the recommended information output procedure, display information for the information terminal of the specific consumer including the recommended information determined for the authenticated specific consumer is output to the information terminal of the specific consumer Is done. Thereby, effective recommendation targeting a specific consumer can be performed on the screen displayed on the information terminal.

  The product recommendation method according to a fifth aspect of the present invention is the product recommendation method according to the fourth aspect, wherein the identification information acquisition procedure relates to the second product displayed on the information terminal of the specific consumer as the specific consumption-related behavior. The identification information of the specific consumer who has performed a predetermined operation on the information screen displaying the display information is acquired for the information, the consumer counting procedure accesses the clustering database, and the identification information The number of specific consumers whose identification information has been acquired in the acquisition procedure is counted for each unit, and the recommended information determination procedure is the specific consumer counted for each unit in the consumer counting procedure. Depending on the number, the information terminal of the at least one consumer associated with the unit is related to the predetermined operation on the information screen And determining the serial recommendation information.

  In the fifth invention of the present application, not only the fact that a direct consumption action is performed on the second product, but also a fact that a predetermined operation such as a click on the information screen, which is a previous stage leading to the consumption action, is performed. The recommended information is also reflected. Thereby, the more effective recommendation which considered the potential consumption demand can be performed.

  The product recommendation method according to a sixth aspect of the present invention is the above-described fifth aspect, wherein the recommended information output procedure is performed on the information terminal of a consumer who has not performed the predetermined operation among the specific consumers. The display information including the recommended information related to the predetermined operation determined in the recommended information determination procedure is output.

  In the sixth invention of the present application, display information including recommended information is output to an information terminal of a specific consumer who does not even perform a predetermined operation such as clicking on information related to the second product on the information screen. This makes it possible to make an effective recommendation targeting a specific consumer who is highly likely not to be aware of the second product at that time.

  In order to achieve the above object, the product recommendation system of the seventh invention associates identification information of a plurality of first products related to the first business with consumption behavior histories of a plurality of consumers for the first product. A first consumer database storing consumption history information for each first product; and history acquisition means for accessing the first consumer database and acquiring consumption history information of a plurality of consumers for a plurality of first products; Clustering information in which each consumer's identification information is associated with one of a plurality of units using the consumption history information of the plurality of consumers for the plurality of first products acquired by the history acquisition means. Clustering means to be generated, identification information of a plurality of second products related to the second business, consumption behavior history of a plurality of consumers for the second product, or specific consumption related other than consumption A second consumer database storing consumption history information associated with a movement history for each second product, and accessing the second consumer database, and for a predetermined second product other than consumption behavior or consumption An identification information acquisition unit that acquires identification information of each of a plurality of consumers who performed a specific consumption-related behavior, a clustering database that stores the clustering information generated by the clustering unit, and an access to the clustering database, According to the consumer counting means for counting the number of consumers whose identification information has been acquired by the identification information acquiring means for each unit, and the number of consumers counted for each unit by the consumer counting means. Information end of at least one of the plurality of consumers associated with the unit To the recommended information determining means for determining recommended information at least related to the consumption behavior or the consumption-related behavior, the recommended information determined by the recommended information determining means is obtained from the recommendation database, the acquired recommended information, Recommended information output means for outputting to the information terminal of at least one consumer.

  In the seventh invention of the present application, identification information of a plurality of first products related to the first business is associated with consumption behavior histories of a plurality of consumers for the first product in advance, and the first consumer database is used as consumption history information. Is remembered. History acquisition means accesses the first consumer database and acquires consumption history information of a plurality of consumers for a plurality of first products. Using the acquired consumption history information, the clustering means generates clustering information by associating each consumer's identification information with one of the plurality of units.

  On the other hand, in the seventh invention of the present application, the identification information of a plurality of second products related to the second business, which is different from the first business, and the consumption behavior history of the plurality of consumers for the second product or a specific The consumption-related behavior history is associated with and stored in the second consumer database as consumption history information. The identification information acquisition means accesses the second consumer database, identifies a plurality of consumers who have performed consumption behavior or consumption related behavior for the predetermined second product, and identifies the plurality of consumers. To get. When consumers are identified in this way, the consumer counting means accesses the clustering database and counts the number of consumers whose identification information has been acquired by the identification information acquiring means for each unit. As a result, the number of consumers who have performed consumption behavior or consumption-related behavior for the specific second product is counted in each unit generated by clustering related to the first product. Thereafter, the recommended information determining means determines recommended information for the information terminal of at least one consumer of the unit according to the count result for each unit. The recommended information determined is acquired from the recommended database by the recommended information output means, and is output to the information terminal of the at least one consumer.

  As a result, for each unit based on the clustering for the first product, recommendation information is determined according to the number of consumers who have performed consumption behavior or consumption-related behavior on the second product, and the determination The recommended information is output to the consumer information terminal of the unit. As a result, even when there is not much consumer consumption history information about the second product, clustering is performed using the first product with abundant consumption history information, and information recommended using the clustering result By determining the contents of, effective recommendations can be made that accurately match the consumer's trend for the second product.

  According to the present invention, even when there is not much consumption history information about a consumer's product, it is possible to make an effective recommendation that accurately matches the consumer's trend for the product.

It is a system configuration figure showing the whole product recommendation system composition of one embodiment of the present invention. It is a table | surface which represents an example of the memory content of a music master database conceptually. It is a table | surface which represents an example of the memory content of a song history database conceptually. It is a table | surface which represents an example of the memory content of a user profile database notionally. It is a table | surface which represents an example of the memory content of a purchase history database conceptually. It is a table | surface which represents an example of the memory content of an article-advertisement database conceptually. It is a table | surface which represents an example of the memory content of a default advertisement database conceptually. It is a table | surface which represents an example of the memory content of the advertisement database classified by user conceptually. It is a table | surface which represents an example of the memory content of an advertisement link database conceptually. It is a table | surface which represents an example of the memory content of an authentication database conceptually. It is explanatory drawing showing an example of my page concerning the said specific consumer displayed on the display connected to PC terminal of a specific consumer. It is a flowchart showing the control procedure regarding clustering which the control part of the PC terminal for calculation performs. It is a flowchart showing the control procedure regarding the determination of the advertising information with respect to a consumer's PC terminal which the control part of the PC terminal for calculation performs. It is a table | surface which represents an example of a UID table notionally. It is a table | surface which represents an example of the consumer number table according to goods conceptually. It is a flowchart showing the detailed procedure of step SA200. It is a flowchart showing the control procedure regarding the output of the advertising information with respect to a consumer's PC terminal which the control part of a WEB server performs. It is a flowchart showing the detailed procedure of step SB100. 10 is a table conceptually showing an example of stored contents of a click history database in a modification in which advertisement information is determined based on click history information. A control procedure related to determination of advertisement information for a consumer's PC terminal, which is executed by the control unit of the PC terminal for calculation, will be described. It is a flowchart showing the detailed procedure of step SA200.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  The overall configuration of the product recommendation system of this embodiment will be described with reference to FIG.

  In FIG. 1, the product recommendation system 1 includes a DB server 100, a calculation PC terminal 200, a WEB server 300, and a PC terminal 400 as an information terminal owned by each consumer.

  The DB server 100 includes a control unit 101, a communication control unit 103, and a storage unit 110.

  The control unit 101 includes a CPU (not shown) and a memory such as a RAM and a ROM. The control unit 101 executes various programs stored in advance in the storage unit 110 while using the temporary storage function of the RAM. As a result, the entire DB server 100 is controlled.

  The communication control unit 103 controls information communication with the calculation PC terminal 200 via a network NW1 such as a local area network (LAN). At the same time, control of information communication with the WEB server 300 via the network NW2 such as a LAN is performed.

  The storage unit 110 is configured by, for example, a hard disk drive (HDD). The storage unit 110 includes an OS storage area 111, an RDBMS storage area 112, and a database storage area 113.

  The OS storage area 111 stores a predetermined operating system (OS). The OS is software that manages the entire computer system.

  The RDBMS storage area 112 stores a predetermined Relational DataBase Management System (RDBMS). RDBMS is software that manages a so-called relational database.

  The database storage area 113 includes a member database (not shown), a music master database 1131 (see FIG. 2 to be described later), a singing history database 1132 as a first consumer database (see FIG. 3 to be described later), and a clustering database. User profile database 1133 (see FIG. 4 described later), purchase history database 1134 as a second consumer database (see FIG. 5 described later), article-advertisement database 1135 (see FIG. 6 described later), default advertisement database 1136 (described later) 7), a user-specific advertisement database 1137 (see FIG. 8 to be described later), an advertisement link database 1138 (see FIG. 9 to be described later), and an authentication database 1139 (see FIG. 10 to be described later) are stored.

  In the member database, member information of a plurality of specific consumers registered as members related to the accompaniment reproduction work of the singing song, in other words, the karaoke work is stored. Member information includes consumer identification information such as UID, gender, date of birth, and the like. The karaoke business corresponds to the first business.

  The DB server 100 having the above configuration is connected to the calculation PC terminal 200 via the network NW1 and is connected to the WEB server 300 via the network NW2.

  A display 220, a keyboard 221, and a mouse 222 are connected to the calculation PC terminal 200. The calculation PC terminal 200 includes a control unit 201, a communication control unit 202, an output control unit 203, an input control unit 204, and a storage unit 210.

  The control unit 201 includes a CPU (not shown) and a memory such as a RAM and a ROM. The control unit 201 executes various programs stored in advance in the storage unit 210 while using the temporary storage function of the RAM. As a result, the entire calculation PC terminal 200 is controlled.

  The communication control unit 202 controls information communication performed with the DB server 100 via the network NW1.

  The output control unit 203 performs control related to the output of the video signal to the display 220. The input control unit 204 performs control related to input of information via the keyboard 221 and the mouse 222.

  The storage unit 210 is composed of, for example, an HDD. The storage unit 210 includes an OS storage area 211 in which a predetermined OS is stored, and a program storage area 212 in which a user profile determination processing program is stored.

  The user profile determination processing program is a program for causing the control unit 201 to execute processing for generating clustering information (described later) stored in the user profile database 1133 of the DB server 100. In the present embodiment, the control unit 201 executes predetermined clustering according to the user profile determination processing program at a predetermined interval, for example, once / month. Note that the clustering is a method of automatically classifying a plurality of data without an external standard using a clustering method such as a well-known K-means method or a Self Organizing Map (SOM) method, or an algorithm thereof. is there.

  On the other hand, the WEB server 300 includes a control unit 301, communication control units 302 and 303, and a storage unit 310.

  The control unit 301 includes a CPU and a memory such as a RAM and a ROM (not shown). The control unit 301 executes various programs stored in advance in the storage unit 310 while using the temporary storage function of the RAM. Thus, the entire WEB server 300 is controlled.

  The communication control unit 302 controls information communication performed with the DB server 100 via the network NW2. The communication control unit 303 controls information communication between the PC terminal 400 and the advertisement distribution server 500 described later via a network NW3 such as a wide area network (WAN).

  The storage unit 310 is composed of, for example, an HDD. The storage unit 310 includes an OS storage area 311 in which a predetermined OS is stored, a web page data storage area 312, and a program storage area 313.

  The web page data storage area 312 stores various types of data such as Hyper Text Markup Language (HTML) data, image data, text data, and the like that make up the web page of the website related to the karaoke business.

  The program storage area 313 stores a WEB server program and an advertisement distribution program.

  The WEB server program is a program for causing the control unit 301 to execute processing for generating web page display information based on various data stored in the web page data storage area 312 and outputting the web page display information. is there.

  The advertisement distribution program is a program for causing the control unit 301 to execute processing for distributing advertisements to be displayed in an advertisement display area (described later) provided on a web page.

  The WEB server 300 having the above configuration is connected to the PC terminal 400 via the network NW3.

  The PC terminal 400 is a PC terminal capable of browsing the web page by accessing the WEB server 300 via the network NW3. A display 420, a keyboard 421, and a mouse 422 are connected to the PC terminal 400. The PC terminal 400 includes a control unit 401, a communication control unit 402, an output control unit 403, an input control unit 404, and a storage unit 410.

  The control unit 401 includes a CPU (not shown) and a memory such as a RAM and a ROM. The control unit 401 executes various programs stored in advance in the storage unit 410 while using the temporary storage function of the RAM. As a result, the entire PC terminal 400 is controlled.

  The communication control unit 402 controls information communication with the WEB server 300 via the network NW3.

  The output control unit 403 performs control related to the output of the video signal to the display 420. The input control unit 404 performs control related to information input via the keyboard 421 and the mouse 422.

  The storage unit 410 is composed of, for example, an HDD. The storage unit 410 includes an OS storage area 411 in which a predetermined OS is stored, and a web browser storage area 412 in which a predetermined web browser is stored.

  The information terminal may be any terminal that can be connected to the WEB server 300 via a network, and may be, for example, a mobile phone or a personal digital assistant (PDA). In the example illustrated in FIG. 1, a configuration in which one consumer's PC terminal 400 is connected to the WEB server 300 is illustrated, but in reality, a plurality of consumer information terminals are connected. Yes.

  The WEB server 300 is connected to the advertisement distribution server 500 via the network NW3.

  The advertisement distribution server 500 is a server that outputs advertisement information as recommended information for recommending a specific product to the WEB server 300 via the network NW3. The advertisement distribution server 500 includes an advertisement database (not shown) as a recommendation database that stores a plurality of advertisement information. The advertisement information is information related to the article such as a product name, product content, price, and appearance image.

  FIG. 2 shows an example of the contents stored in the music master database 1131.

  As shown in FIG. 2, in the music master database 1131, music IDs that are identification information of a plurality of singing songs and singer IDs that are singer identification information corresponding to the singing music are associated with the karaoke service. Attached and memorized. Note that the song music (hereinafter abbreviated as “music”) corresponds to the first product.

  FIG. 3 shows an example of the contents stored in the singing history database 1132.

  As shown in FIG. 3, in the singing history database 1132, singing history information in which music IDs of a plurality of songs related to the karaoke work and a plurality of consumer singing histories for the music are associated with each song Is remembered. The singing history corresponds to the consumption behavior history. Singing history information corresponds to consumption history information.

  The singing history information includes the UID of the consumer who performed the singing on the music related to the karaoke service, the singing date and time information on the date and time when the tune was sung, and the tune ID of the tune. Note that the fact that the consumer has sung the music, that is, that the consumer has sung the music corresponds to the consumer performing a consumption action on the music.

  FIG. 4 shows an example of the contents stored in the user profile database 1133.

  As shown in FIG. 4, the user profile database 1133 stores clustering information generated by clustering executed by the control unit 201 of the calculation PC terminal 200. In the clustering information, each consumer's UID is associated with one of a plurality of units (specifically, unit IDs that are identification information of a plurality of units; the same applies hereinafter) having different tastes or consumption behavior periods by clustering. Information. Note that the number of the units is determined in advance.

  That is, in the user profile database 1133, UIDs of a plurality of consumers and unit IDs of units to which the consumers belong are associated and stored. In this embodiment, based on the singing history information of a plurality of consumers for a plurality of songs stored in the singing history database 1132 of the DB server 100, the UIDs of the plurality of consumers are clustered into a plurality of units. (Details will be described later).

  FIG. 5 shows an example of the contents stored in the purchase history database 1134.

  As shown in FIG. 5, the purchase history database 1134 includes identification information of a plurality of items related to the item sales business (hereinafter referred to as “EC sales business” as appropriate) in Electronic Commerce (EC). Purchase history information in which a certain article ID is associated with purchase histories of a plurality of consumers for the article is stored for each article. The EC sales business corresponds to the second business. The article corresponds to the second product. The purchase history corresponds to the consumption behavior history. The purchase history information corresponds to consumption history information.

  The purchase history information includes the UID of the consumer who purchased the item related to the EC sales business, the purchase date and time information indicating the date and time when the item was purchased, and the item ID of the item. Note that the purchase of an article by the consumer corresponds to the consumer performing a consumption action on the article.

  In this example, the purchase history database 1134 is constructed by automatically storing purchase history information based on purchase histories of a plurality of consumers for a plurality of articles related to the EC sales business. However, the present invention is not limited to this, and in the purchase history database 1134, an employee of the store manually inputs purchase history information based on purchase history of a plurality of consumers for a plurality of items related to sales of the item at the store. It may be constructed by.

  FIG. 6 shows an example of the contents stored in the article-advertisement database 1135.

  As shown in FIG. 6, in the article-advertisement database 1135, article IDs of a plurality of articles related to the EC sales business and advertisement IDs that are identification information of advertisement information related to the articles are stored in association with each other. Yes.

  FIG. 7 shows an example of the contents stored in the default advertisement database 1136.

  As shown in FIG. 7, the default advertisement database 1136 stores advertisement IDs of advertisement information related to predetermined specific articles, such as newly released articles and recommended articles.

  FIG. 8 shows an example of the contents stored in the user-specific advertisement database 1137.

  As illustrated in FIG. 8, the user-specific advertisement database 1137 includes a plurality of consumer UIDs and advertisement IDs of advertisement information for the consumer PC terminal 400 determined by the control unit 201 of the calculation PC terminal 200. Are stored in association with each other. A method for determining advertisement information for each consumer's PC terminal 400 will be described in detail with reference to FIGS.

  In addition, you may make it memorize | store advertisement information itself, ie, the information itself regarding goods, such as a brand name, goods content, a price, an external appearance image, for example in this advertisement database classified by user 1137.

  FIG. 9 shows an example of the contents stored in the advertisement link database 1138.

  As shown in FIG. 9, the advertisement link database 1138 includes an advertisement ID of a plurality of advertisement information, and a Uniform Resource Locator (URL) that specifies a position in the advertisement database of the advertisement distribution server 500 that stores the advertisement information. Are stored in association with each other.

  FIG. 10 shows an example of the contents stored in the authentication database 1139.

  As shown in FIG. 10, the authentication database 1139 previously stores a plurality of specific consumer authentication information regarding the karaoke business. The authentication information is given, for example, when a consumer is registered as a member related to karaoke business. This authentication information includes an authentication ID and a password.

  In this example, all the databases shown in FIGS. 2 to 10 described above are stored in the storage unit 110 of the DB server 100, but the present invention is not limited to this. That is, these databases may be separately stored in the storage unit 210 of the calculation PC terminal 200, the storage unit 110 of the DB server 100, the storage unit 310 of the WEB server 300, and the like.

  In the present embodiment, the consumer activates the web browser stored in the web browser storage area 412 of the PC terminal 400 when he / she wants to browse a web page of a website having contents related to the karaoke business. Then, the keyboard 421 and the mouse 422 connected to the PC terminal 400 are operated to input the web page of the website related to the karaoke business, for example, the URL of the top page of the website. Then, a request is output from the consumer's PC terminal 400 to the WEB server 300.

  When the web server 300 receives a request from the consumer's PC terminal 400, the top page display information is generated in response to the request, and the display information is output to the PC terminal 400 related to the request. Is done. By this. On the display 420 connected to the PC terminal 400, the top page of the website with contents related to karaoke work is displayed.

  Although not shown, the top page includes the above-described authentication information, that is, an authentication ID and password input box. The consumer operates the keyboard 421 and the mouse 422 to input authentication information into the input box when he / she wants to browse a My Page that is a member-only web page related to the karaoke business. Then, an authentication request signal including the input authentication information is output from the consumer's PC terminal 400 to the WEB server 300.

  When the authentication request signal from the consumer's PC terminal 400 is received by the WEB server 300, the authentication database 1139 of the DB server 100 is accessed based on the authentication request signal, and the consumer related to the authentication request signal is accessed. Authentication is performed (details will be described later). At this time, when the consumer related to the authentication request signal is authenticated, the advertisement ID of the advertisement information for the PC terminal 400 of the authenticated consumer in the WEB server 300 is obtained from the advertisement database for each user 1137 of the DB server 100. To be acquired. Then, advertisement information corresponding to the acquired advertisement ID is acquired from the advertisement database of the advertisement distribution server 500. Then, the information screen for the authenticated consumer's PC terminal 400 including the acquired advertisement information, that is, the information for the specific consumer's PC terminal 400 in which the authentication information is stored in the authentication database 1139. Display information of the screen is generated. Hereinafter, the information screen for a specific consumer's PC terminal 400 will be referred to as “my page related to a specific consumer”, “my page”, or the like.

  Thereafter, the generated my page display information related to the specific consumer is output from the WEB server 300 to the PC terminal 400 of the specific consumer related to the authentication request signal. When the display information of My Page is received at the PC terminal 400 of the specific consumer, the display screen of the display 420 connected to the PC terminal 400 of the specific consumer displays the display information. Transition to My Page related to a specific consumer.

  FIG. 11 shows an example of a My Page related to a specific consumer displayed on the display 420 connected to the PC terminal 400 of the specific consumer.

  As shown in FIG. 11, this my page includes a plurality of target advertisement display areas TA, a common advertisement display area CA, and a member content display area MA for displaying member contents.

  The target advertisement display area TA is an advertisement display area in which an advertisement targeting a specific consumer related to My Page is displayed. In this example, an advertisement displayed as text “advertisement 1” is displayed in the target advertisement display area TA1. In the target advertisement display area TA2, an advertisement having a text “Advertisement 2” is displayed.

  The common advertisement display area CA is an advertisement display area in which a predetermined advertisement is displayed. In this example, in the common advertisement display area CA, an advertisement displayed as text “advertisement 3” is displayed.

  Further, when a click on the advertisement displayed in the target advertisement display area TA or the common advertisement display area CA on the My Page, in other words, information related to an article is performed via the mouse 422, the WEB server 300 is displayed. The click history is output at. At the same time, the display screen of the display 420 transitions to a web page related to an article related to the clicked advertisement, such as a page for purchasing the article or a page for posting more detailed information on the article. To do. The click through the mouse 422 corresponds to a predetermined operation.

  Hereinafter, a control procedure according to the product recommendation method in this embodiment, which is executed by the calculation PC terminal 200, the WEB server 300, and the like will be described with reference to FIGS. 12, 13, 16, 17, and 18. FIG.

  A control procedure related to clustering executed by the control unit 201 of the calculation PC terminal 200 will be described with reference to FIG.

  In FIG. 12, for example, when the operator inputs a predetermined processing start command using the keyboard 221 or the mouse 222, this flow is started as represented by the “START” position in the figure. .

  First, in step SA10, the control unit 201 accesses the singing history database 1132 (see FIG. 3) of the DB server 100 via the communication control unit 202 and the network NW1. And the singing log | history information of several consumers with respect to several music is acquired. This step corresponds to the history acquisition procedure described in each claim and functions as history acquisition means.

  Thereafter, in step SA20, the control unit 201 accesses the music master database 1131 (see FIG. 2) of the DB server 100 via the communication control unit 202 and the network NW1. And all singer ID matched with the said music ID is acquired by using as a key all the music ID contained in the some consumer's singing history information with respect to the several music acquired by said step SA10.

  And it moves to step SA30 and the control part 201 was acquired by said step SA20 about each of the said some consumers using the song history information of the some consumer with respect to the some music acquired by said step SA10. A feature vector for the singer's music corresponding to all singer IDs is generated. And the feature vector produced | generated about each of several consumers is matched with the said consumer's UID.

  It should be noted that the feature vectors generated for each of the plurality of consumers in this step are vectors in which all the singer IDs obtained in step SA20 are the elements of the vector and the value of the element is “1” or “0”. It is. An element value of “1” means that there is a singing history for the song of the singer corresponding to the singer ID related to the element, and an element value of “0” means a singer ID related to the element. This means that there is no singing history for the singer's corresponding song.

  For example, for a consumer who has sung the singer's music corresponding to all the singer IDs acquired in step SA20, a feature vector whose element values are all “1” is generated. That is, the consumer feature vector in this case is <1,1,1,1,1. On the other hand, for a consumer who has not sung the singer's music corresponding to all the singer IDs acquired in step SA20, a feature vector whose element values are all “0” is generated. That is, the consumer feature vector in this case is <0, 0, 0, 0, 0.

  Therefore, the feature vector indicates whether the singer's music has been sung or not sung for all the singer corresponding to the singer ID acquired in step SA20. In other words, the feature vector represents the consumer's singer's preference.

  Thereafter, in step SA40, the control unit 201 uses a plurality of feature vectors respectively associated with the UIDs of the plurality of consumers in step SA30 using the initial values of the representative vectors of the plurality of units prepared in advance. Are classified into a plurality of units by a clustering method such as a known K-means method or SOM method. Corresponding to the result of the clustering, the UID of each consumer is associated with the unit ID of the unit into which the feature vector corresponding to the UID is classified to generate clustering information.

  In step SA50, the control unit 201 accesses the user profile database 1133 (see FIG. 4) of the DB server 100 via the communication control unit 202 and the network NW1. Then, the clustering information generated in step SA40 is stored in the user profile database 1133. If the clustering information is already stored in the user profile database 1133, the contents are updated to the contents of the newly generated clustering information. Thereafter, this flow is terminated.

  In the above, step SA30 and step SA40 correspond to the clustering procedure described in each claim and function as clustering means.

  A control procedure related to determination of advertisement information for the consumer's PC terminal 400, which is executed by the control unit 201 of the calculation PC terminal 200, will be described with reference to FIG.

  In FIG. 13, for example, when the operator inputs a predetermined processing start command using the keyboard 221 or the mouse 222, this flow is started as represented by the “START” position in the figure. .

  First, in step SA105, the control unit 201 acquires information related to a period range that is preset by an operator and stored in an appropriate area of the storage unit 213, for example. A period range represented by the acquired information is defined as D. The period range = D is set as “2009/01/01 to 2009/12/31”, for example. The period range is not limited to “year / month / day” units, and may include time units.

  Thereafter, in step SA110, the control unit 201 sets the value of the variable U corresponding to any of the plurality of units to 1.

  In step SA115, the control unit 201 accesses the user profile database 1133 (see FIG. 4) of the DB server 100 via the communication control unit 202 and the network NW1. Then, all UIDs associated with the unit ID are acquired using the unit ID of the unit corresponding to the value of the variable U at this time as a key. That is, in this step, the control unit 201 obtains the UIDs of a plurality of consumers belonging to the unit corresponding to the value of the variable U at this time. That is, in this step, the UIDs of a plurality of consumers belonging to the unit are acquired for each unit.

  Thereafter, in step SA120, the control unit 201 tabulates the acquisition result in step SA115 and stores it as a UID table (see FIG. 14 described later) in a memory such as a RAM in the control unit 101 of the DB server 100, for example. Let

  Then, the process proceeds to step SA125. The control unit 201 detects the number of records in the UID table generated in step SA120. The number of detected records is N. A “record” is a unit of data constituting a table or a database. That is, the number of records detected in this step = N matches the number of consumers belonging to the unit corresponding to the value of the variable U at the time when step SA115 is executed.

  Thereafter, the process proceeds to step SA130, and the control unit 201 accesses the purchase history database 1134 (see FIG. 5) of the DB server 100 via the communication control unit 202 and the network NW1. The purchase date / time is included in all purchase history information within the period range set at step SA105 = D and the UID matches one of the UIDs in the UID table generated at step SA120. The UID and the article ID are acquired in association with each other. That is, in this step, the control unit 201 purchases the article sold within the period range set in step SA105 as the predetermined article = D, and the variable at the time when step SA115 is executed. The UID of each of the plurality of consumers belonging to the unit corresponding to the value of U is acquired in association with the item ID of the item purchased by the consumer. This step corresponds to the identification information acquisition procedure described in each claim and functions as an identification information acquisition means.

  Then, the process proceeds to step SA135, where the control unit 201 classifies all the UIDs acquired in step SA130 for each corresponding article ID, and in other words, the consumption in which the UID is acquired in step SA130. The number of persons is counted for each article. That is, in this step, the control unit 201 corresponds to the value of the variable U at the time when the step SA115 is executed for each article sold within the period range set at the step SA105 = D. The number of consumers who purchased the article in the unit, that is, the number of purchasers is counted. That is, in this step, the number of consumers whose UID has been acquired in step SA130 is counted for each unit.

  After that, in step SA140, the control unit 201 tabulates the count result in step SA135, and stores it as an article-specific consumer number table (see FIG. 15 described later), such as a RAM in the control unit 101 of the DB server 100, for example. Store in the memory.

  Then, the process proceeds to step SA145, and the control unit 201 acquires information related to the number of articles to be advertised that are set in advance by the operator and stored in an appropriate area of the storage unit 213, for example. The number of articles to be advertised represented by the acquired information is C.

  After that, in step SA150, the control unit 201 sets the value of the variable i for counting the number of articles to be advertised to 1 (one).

  Then, the process proceeds to step SA155, where the control unit 201 obtains an article ID included in the data in which the number of consumers counted for each article in step SA135 is i-th from the article-specific consumer number table. . That is, in this step, the control unit 201 obtains the article ID of the article having the i-th largest number of purchasers in the unit corresponding to the value of the variable U at the time when the step SA115 is executed. . The acquired article ID is I.

  Thereafter, in step SA200, the control unit 201 determines, according to the number of consumers counted in step SA135, the PCs of consumers belonging to the unit corresponding to the value of the variable U when step SA115 is executed. An advertisement determination process for determining advertisement information for the terminal 400 is executed. This detailed content will be described later with reference to FIG.

  Then, the process proceeds to step SA160, where the control unit 201 adds 1 to the value of the variable i, and the process proceeds to step SA165.

  In step SA165, the control unit 201 determines whether the value of the variable i is larger than the number of articles to be advertised = C set in step SA145. If i ≦ C, the determination is not satisfied and the routine goes to Step SA155 to repeat the same procedure. On the other hand, if i> C, the determination is satisfied and the routine goes to Step SA170.

  In step SA170, the control unit 201 adds 1 to the value of the variable U, and proceeds to step SA175.

  In step SA175, the control unit 201 determines whether or not the value of the variable U is greater than a predetermined number M of units. In other words, it is determined whether or not the procedure after step SA115 is executed for all units for each unit. If U ≦ M, the determination is not satisfied and the routine returns to Step SA115 and the same procedure is repeated. On the other hand, if U> M, the determination is satisfied and this flow is terminated.

  In the above, step SA115 and step SA135 correspond to the consumer counting procedure described in each claim and function as a consumer counting means.

  FIG. 14 shows an example of the UID table. As shown in FIG. 14, the UID table stores the UIDs of consumers belonging to the unit corresponding to the value of the variable U at the time when step SA115 in FIG. 13 is executed.

  FIG. 15 shows an example of the consumer number table by item. As shown in FIG. 15, the article-specific consumer number table stores, for each article, the article ID of the article and the number of consumers counted in step SA135 in FIG.

  The detailed procedure of step SA200 in FIG. 13 will be described with reference to FIG.

  In FIG. 16, first in step SA205, the control unit 201 sets the value of a variable j for counting the number of records in the UID table to 1.

  Thereafter, in step SA210, the control unit 201 acquires one UID stored in the UID table. The acquired UID is P.

  Then, the process proceeds to step SA215, and the control unit 201 deletes UID = P acquired in step SA210 from the UID table.

  Thereafter, in step SA220, the control unit 201 accesses the purchase history database 1134 of the DB server 100 via the communication control unit 202 and the network NW1. Then, the presence / absence of purchase history information in which the UID matches the UID = P acquired in step SA210 and the item ID matches the item ID = I acquired in step SA155 of FIG. 13 is determined. In other words, it is determined whether or not the consumer corresponding to UID = P acquired in step SA210 has already purchased the item corresponding to article ID = I acquired in step SA155 of FIG. . If the purchase history information is present, in other words, if the consumer corresponding to UID = P has already purchased the item corresponding to item ID = I, the determination is satisfied, and the process goes to step SA240 described later. Move. On the other hand, if there is no purchase history information, in other words, if the consumer corresponding to UID = P has not purchased the item corresponding to item ID = I, the determination is not satisfied, and the routine goes to Step SA225.

  Then, the process proceeds to Step SA225, and the control unit 201 accesses the article-advertisement database 1135 (see FIG. 6) of the DB server 100 via the communication control unit 202 and the network NW1. Then, the advertisement ID associated with the article ID is acquired using the article ID = I acquired in step SA155 of FIG. 13 as a key. The acquired advertisement ID is B.

  Thereafter, in step SA230, the control unit 201 accesses the user-specific advertisement database 1137 (see FIG. 8) of the DB server 100 via the communication control unit 202 and the network NW1. Then, the advertisement ID = B acquired in step SA225 is stored in the user-specific advertisement database 1137 in association with the UID = P acquired in step SA210, and the advertisement information corresponding to the advertisement ID = B is stored. It is determined as advertisement information for the consumer's PC terminal 400 corresponding to UID = P.

  In step SA155 of FIG. 13, the article IDs of the top C articles having a large number of purchasers in the unit corresponding to the value of variable U at the time when step SA115 is executed are acquired. . Therefore, in this step SA230, the control unit 201 displays advertisement information regarding an article having a large number of purchased consumers, specifically, a top C article having a large number of purchased consumers in the unit. A decision is made to give priority to advertisement information related to goods with a small number of purchased consumers.

  Then, the process proceeds to step SA235, where the control unit 201 adds 1 to the value of the variable j, and then proceeds to step SA240.

  In step SA240, the control unit 201 determines whether or not the value of the variable j is greater than the number of records in the UID table detected in step SA125 of FIG. If j ≦ N, the determination is not satisfied and the routine returns to step SA210 and the same procedure is repeated. On the other hand, if j> N, the determination is satisfied and the routine is terminated.

  As described above, in step SA200, the control unit 201 associates with the unit according to the number of consumers counted for each unit in step SA135 in FIG. The advertisement information related to the purchase behavior of the goods with respect to the PC terminal 400 of at least one of the plurality of consumers belonging to 1 is determined. Therefore, all the procedures of Step SA200 shown in FIG. 16, that is, Steps SA205 to SA240 correspond to the recommended information determination procedure described in each claim and function as recommended information determination means.

  The control procedure regarding the output of advertisement information to the consumer's PC terminal 400 executed by the control unit 301 of the WEB server 300 will be described with reference to FIG.

  In FIG. 17, for example, when a request related to the top page of the website related to the karaoke business is input from the PC terminal 400 of the consumer, this flow starts as represented by the “START” position in the figure. Is done.

  First, in step SB10, the control unit 301 receives various requests stored in the web page data storage area 312 in response to a request related to the top page received from the consumer's PC terminal 400 via the network NW3 and the communication control unit 303. Display information of the top page is generated based on the data. Then, the generated display information of the top page is output to the consumer PC terminal 400 related to the request via the communication control unit 303 and the network NW3. As a result, the top page displaying the display information is displayed on the display 420 connected to the PC terminal 400 of the consumer.

  Thereafter, in step SB20, the control unit 301 determines whether or not the authentication request signal described above has been received from the consumer's PC terminal 400 via the network NW3 and the communication control unit 303. Until the authentication request signal is received, the determination is not satisfied and the process waits in a loop. When the authentication request signal is received from the consumer's PC terminal 400, the determination is satisfied and the routine goes to Step SB30. When the consumer is a specific consumer, an authentication request signal from the PC terminal 400 of the specific consumer is received. This step corresponds to the authentication request receiving procedure described in each claim.

  In step SB30, the control unit 301 accesses the authentication database 1139 (see FIG. 10) of the DB server 100 via the communication control unit 302 and the network NW2. Then, it is determined whether the authentication information included in the authentication request signal received in step SB20 matches any of a plurality of specific consumer authentication information stored in the authentication database 1139. As a result, authentication of a specific consumer related to the authentication request signal is performed. This step corresponds to the authentication procedure described in each claim. If the authentication information included in the authentication request signal does not match any of the plurality of specific consumer authentication information stored in the authentication database 1139, the authentication is not performed, so the determination is not satisfied, and the routine goes to Step SB40.

  In step SB40, the control unit 301 generates display information of a predetermined error screen based on various data stored in the web page data storage area 312. The generated error screen display information is output to the consumer PC terminal 400 related to the authentication request signal received in step SB20 via the communication control unit 303 and the network NW3. As a result, an error screen displaying the display information is displayed on the display 420 connected to the PC terminal 400 of the consumer. Thereafter, this flow is terminated.

  On the other hand, in step SB30, if the authentication information included in the authentication request signal matches any of a plurality of specific consumer authentication information stored in the authentication database 1139, authentication is performed. Is satisfied, and the routine goes to Step SB100.

  In step SB100, the control unit 301 displays, on the PC terminal 400 of the specific consumer, my page display information related to the specific consumer, including advertisement information for the specific consumer authenticated in step SB20. The advertisement output process to be output is executed. This detailed content will be described later with reference to FIG. Thereafter, this flow is terminated.

  The detailed procedure of step SB100 in FIG. 17 will be described with reference to FIG.

  18, first, in step SB105, the control unit 301 acquires the UID of the specific consumer authenticated in step SB20 in FIG. 17 based on the authentication request signal received in step SB10 in FIG. The acquired UID is X.

  Thereafter, in step SB110, the control unit 301 acquires information related to the number of target advertisement display areas TA in the above-described my page, which is set in advance by an operator and stored in an appropriate area of the storage unit 313, for example. The number of target advertisement display areas TA in the My Page represented by the acquired information is Y.

  In step SB115, the control unit 301 accesses the user-specific advertisement database 1137 (see FIG. 8) of the DB server 100 via the communication control unit 302 and the network NW2. Then, using the UID = X acquired in step SB105 as a key, Y advertisement IDs associated with the UID are acquired. That is, the control unit 301 obtains Y advertisement IDs of the advertisement information determined in step SA200 in FIG. 13 for the specific consumer corresponding to UID = X authenticated in step SB20 in FIG. . In addition, since the number of advertisement IDs of advertisement information determined for a specific consumer corresponding to UID = X is less than Y in the user-specific advertisement database 1137, Y advertisement IDs associated with the UID are acquired. Sometimes it is not possible. Then, the control unit 301 lists the advertisement IDs acquired from the user-specific advertisement database 1137 and stores the list as an advertisement ID list in, for example, a memory such as a RAM in the control unit 101 of the DB server 100.

  In step SA220 to step SA230 in FIG. 16 above, for each unit, it is determined whether the consumer has purchased an article with a large number of purchasers in the unit, and the consumer who has not purchased the article has been purchased. The UID is stored in the user-specific advertisement database 1137 in association with the advertisement ID of the advertisement information related to the article. Then, in this step SB115, Y advertisement IDs of advertisement information relating to articles not purchased by the consumer are acquired. Therefore, in step SB170, which will be described later, the display information including the advertising information determined in step SA200 of FIG. 13 for each unit belongs to the unit, and the PC of the consumer who has not purchased an article related to the advertising information. It is output to the terminal 400.

  Thereafter, in step SB120, the control unit 301 detects the number of advertisement IDs included in the advertisement ID list generated in step SB115. The number of advertisement IDs detected is Z. That is, the value of Z is an integer from 0 to Y.

  Then, the process proceeds to step SB125, and the control unit 301 sets the value of the variable k for counting the number of advertisement IDs included in the advertisement ID list to 1.

  Thereafter, the process proceeds to step SB130, and the control unit 301 determines whether the value of the variable k is equal to or less than the number of advertisement IDs detected in step SB120 = Z. In other words, it is determined whether all advertisement IDs included in the advertisement ID list have been acquired in step SB135 described later. If k ≦ Z, in other words, if all advertisement IDs included in the advertisement ID list have not yet been acquired in step SB135, which will be described later, the determination is satisfied and the routine goes to step SB135.

  In step SB135, the control unit 301 acquires the kth advertisement ID included in the advertisement ID list generated in step SB115. The acquired advertisement ID is A. Thereafter, the process proceeds to step SB145.

  On the other hand, if k> Z in step SB130, in other words, if all advertisement IDs already included in the advertisement ID list have already been acquired in step SB135, the determination in step SB130 is not satisfied and step SB140 is satisfied. Move on.

  In step SB140, the control unit 301 accesses the default advertisement database 1136 (see FIG. 7) of the DB server 100 via the communication control unit 302 and the network NW2. Then, the advertisement ID of one advertisement information among the advertisement IDs of the plurality of advertisement information stored in the default advertisement database 1136 is acquired at random. The acquired advertisement ID is A.

  Then, the process proceeds to step SB145, and the control unit 301 accesses the advertisement link database 1138 (see FIG. 9) of the DB server 100 via the communication control unit 302 and the network NW2. Then, the URL associated with the advertisement ID is acquired using the advertisement ID = A acquired in step SB135 or SB140 as a key. The acquired URL is set to L.

  Thereafter, in step SB150, the control unit 301 accesses the advertisement database of the advertisement distribution server 500 via the communication control unit 303 and the network NW3. Then, the URL = L acquired in step SB145 is specified, and the advertisement information stored at the location of the URL is acquired.

  Then, the process proceeds to step SB155, and the control unit 301 adds 1 to the value of the variable k, and the process proceeds to step SB160.

  In step SB160, the control unit 301 determines whether or not the value of the variable k is larger than the number of target advertisement display areas TA in the my page = Y. In other words, it is determined whether or not Y pieces of advertisement information are acquired from the advertisement database of the advertisement distribution server 500. When k ≦ Y, in other words, when Y pieces of advertisement information have not been acquired from the advertisement database of the advertisement distribution server 500, the determination is not satisfied and the process returns to step SB130 and the same procedure is repeated. On the other hand, if k> Y, in other words, if Y pieces of advertisement information have already been acquired from the advertisement database of the advertisement distribution server 500, the determination is satisfied and the routine goes to Step SB165.

  In step SB165, the control unit 301 converts the advertisement information acquired in step SB150 to an HTML format. Then, display information of my page related to a specific consumer related to UID = X acquired in step SB105, in which the advertisement information in the HTML format is incorporated in the target advertisement display area TA, is generated.

  Thereafter, in step SB170, the control unit 301 transmits the display information of the my page related to the specific consumer corresponding to UID = X generated in step SB165, via the communication control unit 303 and the network NW3. Output to the personal computer terminal 400. In other words, the display information including the advertisement information determined in step SA200 of FIG. 13 obtained from the user-specific advertisement database 1137 is output to the PC terminal 400 of one consumer. Equivalent to. This step corresponds to the recommended information output procedure described in each claim and functions as a recommended information output means. As a result, the My Page related to the specific consumer displaying the display information is displayed on the display 420 connected to the PC terminal 400 of the specific consumer corresponding to UID = X. Then, this routine ends.

  As described above, in this embodiment, singing history information related to karaoke work is stored in advance in the singing history database 1132 of the DB server 100 (see FIG. 3). The control unit 201 of the calculation PC terminal 200 accesses the singing history database 1132 and acquires singing history information of a plurality of consumers for a plurality of songs. Then, using the acquired singing history information, the UID of each consumer is associated with one of a plurality of units to generate clustering information.

  On the other hand, in the present embodiment, purchase history information related to EC sales work different from the karaoke work is stored in advance in the purchase history database 1134 (see FIG. 5) of the DB server 100. The control unit 201 accesses the purchase history database 1134, specifies a plurality of consumers who have purchased a predetermined article, and acquires the UIDs of the plurality of consumers. When the consumer is specified in this way, the control unit 201 accesses the user profile database 1133 (see FIG. 4) of the DB server 100, and determines the number of consumers for which the UID has been acquired for each unit. Count. As a result, the number of consumers who have purchased a specific article is counted in each unit generated by clustering for music. Thereafter, the control unit 201 determines advertisement information for the PC terminal 400 of the consumer of the unit according to the count result for each unit (see step SA200 in FIG. 13). Then, the control unit 301 of the WEB server 300 acquires the determined advertisement information from the user-specific advertisement database 1137 (see FIG. 8) of the DB server 100 and outputs it to the PC terminal 400 of the consumer. (See step SB170 in FIG. 18).

  As a result, advertisement information corresponding to the number of consumers who have purchased the item is determined for each unit based on the clustering related to music, and is output to the PC terminal 400 of the consumer of the unit. As a result, even if there is not much consumption history information about consumer goods, that is, purchase history information, clustering is performed using abundant consumption history information about consumer songs, that is, singing history information. By determining the content of the recommended information using the clustering result, it is possible to make an effective recommendation that accurately matches the consumer's trend for the article.

  Further, in this embodiment, in particular, the control unit 301 of the WEB server 300 performs an advertisement related to an article with a large number of purchased consumers for each unit in step SA200 of FIG. 13 (see FIG. 16 for details). A decision is made to prioritize information over advertising information relating to goods with a small number of purchased consumers. As a result, for each unit, the advertisement information for an article with a large number of purchased consumers is determined in preference to the advertisement information for other articles. As a result, it is possible to make an effective recommendation reflecting the fact that a relatively large number of consumers in the unit purchase the item.

  In this embodiment, in particular, the control unit 301 of the WEB server 300 associates the advertisement information determined in step SA200 of FIG. 13 with each unit in step SB170 of FIG. It outputs to the PC terminal 400 of the consumer who has not purchased the goods regarding the said advertisement information. Thereby, in each unit, advertisement information about an article purchased by a relatively large number of consumers is output to the PC terminal 400 of a consumer who has not purchased the article. As a result, it is possible to make a recommendation targeting consumers who have not purchased the goods at that time. That is, finally, it is possible to reliably make an effective recommendation so that all consumers in one unit having the same preference complete the purchase of the item.

  In the present embodiment, in particular, authentication information of a specific consumer is stored in advance in the authentication database 1139 (see FIG. 10) of the DB server 100 regarding the karaoke business. The control unit 301 of the WEB server 300 receives an authentication request signal from the PC terminal 400 of a specific consumer (see step SB10 in FIG. 17). Based on the received authentication request signal, the authentication database 1139 is accessed to authenticate a specific consumer related to the authentication request signal (see step SB20 in FIG. 17). Then, the display information of the My Page related to the specific consumer including the advertisement information determined in step SA200 of FIG. 13 regarding the authenticated specific consumer is displayed on the PC terminal 400 of the specific consumer. Output. Thereby, effective recommendation targeting a specific consumer can be performed on the screen displayed on the PC terminal 400.

  The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and technical idea of the present invention. Hereinafter, such modifications will be described.

(1) When advertising information is determined based on click history information In the above embodiment, advertising information is determined based on purchase history information of a plurality of consumers for a plurality of articles, but the present invention is not limited thereto. . In other words, the advertisement information may be determined based on the click history information on information related to a plurality of articles of a plurality of consumers.

  For example, a specific consumer who is browsing the above-mentioned my page (see FIG. 11) relates to the content of the advertisement displayed on the target advertisement display area TA or the common advertisement display area CA on the my page, that is, the article. If you are interested in information, you may click on the advertisement to view more detailed information. In this modification, the control unit 301 of the WEB server 300 aggregates the click history for the advertisement displayed on the my page, in other words, the information related to the article into a database, and uses the DB server as a click history database. 100 database storage areas 113 are stored.

  FIG. 19 shows an example of the contents stored in the click history database.

  As shown in FIG. 19, in the click history database 1140, the click history information in which the article IDs of a plurality of articles related to the EC sales business are associated with the click histories of a plurality of consumers for the article is stored in each article. Every one is remembered. The click history database 1140 corresponds to a second consumer database. The click history corresponds to a specific consumption-related behavior history other than consumption. The click history information corresponds to consumption history information.

  The click history information includes the UID of a specific consumer who made the click on the My Page for the advertisement displayed on the display 420, in other words, the information related to the article, and the date and time information on the click. A certain click date and time and an article ID of an article related to the advertisement are included. Note that when the consumer clicks on the My Page for the advertisement displayed on the display 420, in other words, the information related to the article, the consumer performs a specific consumption-related action other than consumption on the article. Is equivalent to

  Here, in this modification, the control procedure related to clustering executed by the control unit 201 of the calculation PC terminal 200 is the same as that in FIG. Hereinafter, a control procedure regarding determination of advertisement information for the consumer's PC terminal 400, which is executed by the control unit 201 of the calculation PC terminal 200 in the present modification, will be described with reference to FIG. FIG. 20 corresponds to FIG. 13 described above. The same steps as those in FIG. 13 are denoted by the same reference numerals and description thereof is omitted.

  20 differs from FIG. 13 described above in that step SA130 ′, step SA135 ′, and step SA140 ′ are provided in place of step SA130, step SA135, and step SA140. That is, in step SA125, if the number of detected records is N, the control unit 201 proceeds to step SA130 ′ provided in place of step SA130.

  In step SA130 ′, the control unit 201 accesses the click history database 1140 (see FIG. 19) of the DB server 100 via the communication control unit 202 and the network NW1. Then, all the click history information whose click date / time is within the period range = D set in step SA105 described above and whose UID matches one of the UIDs in the UID table generated in step SA120 described above. The UID and the article ID included are acquired in association with each other. That is, in this step, the control unit 201 performs the above-described click on the advertisement that has been posted within the period range = D set in step SA105 described above as a predetermined article, that is, information related to the article. In addition, the UIDs of a plurality of specific consumers belonging to the unit corresponding to the value of the variable U at the time when the above-described step SA115 is executed are the item IDs of the items related to the advertisement clicked by the specific consumer. Acquired in association with. This step corresponds to the identification information acquisition procedure described in each claim and functions as an identification information acquisition means.

  Thereafter, in step SA135 ′ provided in place of step SA135, the control unit 201 classifies all the UIDs acquired in step SA130 ′ for each corresponding article ID, and the number of UIDs, in other words, the above steps. The number of specific consumers whose UID has been acquired at SA 130 'is counted for each article. In other words, in this step, the control unit 201 performs the above-described step SA115 for each article related to the advertisement that has been placed in the period range = D set in the above-described step SA105, that is, information related to the article. The number of consumers who have clicked on the information related to the article in the unit corresponding to the value of the variable U at the time of execution is counted. That is, in this step, the number of consumers whose UID has been acquired in step SA130 ′ is counted for each unit.

  Then, in step SA140 ′ provided in place of step SA140, the control unit 201 tabulates the count result in step SA135 ′, and stores it as an article-specific consumer number table, for example, in the control unit 101 of the DB server 100. It is stored in a memory such as a RAM.

  Subsequent steps after step SA145 are substantially the same as those in FIG.

  In the above, step SA115 and step SA135 ′ correspond to the consumer counting procedure described in each claim and function as a consumer counting means.

  The detailed procedure of step SA200 in the present modification will be described with reference to FIG. FIG. 21 corresponds to FIG. 16 described above. Steps equivalent to those in FIG. 16 are denoted by the same reference numerals and description thereof is omitted.

  FIG. 21 differs from FIG. 16 described above in that step SA220 ′ is provided instead of step SA220. That is, when the control unit 201 deletes UID = P from the UID table in step SA215, the process proceeds to step SA220 ′ provided in place of step SA220.

  In step SA220 ′, the control unit 201 accesses the click history database 1140 of the DB server 100 via the communication control unit 202 and the network NW1. Then, it is determined whether or not there is click history information in which the UID matches the UID = P acquired in step SA210 and the article ID matches the article ID = I acquired in step SA155 of FIG. . In other words, whether the specific consumer corresponding to UID = P acquired in step SA210 described above has clicked on information related to the item corresponding to article ID = I acquired in step SA155 of FIG. Judgment is made. If the click history information is present, in other words, if a specific consumer corresponding to UID = P has already clicked on information related to the item corresponding to item ID = I, the determination is satisfied, Control goes to step SA240. On the other hand, if there is no click history information, in other words, if a specific consumer corresponding to UID = P has not clicked on information related to the item corresponding to item ID = I, the determination is not satisfied, Control goes to step SA225.

  The subsequent steps after step SA225 are substantially the same as those in FIG.

  In step SA200 in this modification, the control unit 201 determines the PC terminal of at least one consumer belonging to the unit according to the number of consumers counted for each unit in step SA135 ′ in FIG. For 400, advertisement information related to the click on the My Page is determined. Accordingly, all the procedures of step SA200 in this modification shown in FIG. 21, that is, steps SA205 to SA215, step SA220 ′, and steps SA225 to SA240 correspond to the recommended information determination procedure described in each claim. , Functioning as recommended information determining means.

  Here, in the present modification, the control procedure related to the output of the advertisement information to the consumer's PC terminal 400, which is executed by the control unit 301 of the WEB server 300, is substantially the same as that shown in FIGS. In step SA220 ′ to step SA230 in FIG. 21 described above, the control unit 201 of the calculation PC terminal 200 has a large number of the specific consumers clicked by the specific consumer for each unit in the unit. It is determined whether or not the information related to the product has been clicked, and the UID of a specific consumer who has not clicked the information related to the product is stored in the user-specific advertising database 1137 in association with the advertising ID of the advertising information related to the product. ing.

  Therefore, the control unit 301 of the WEB server 300 obtains Y advertisement IDs of advertisement information related to information related to the article that the specific consumer has not clicked in step SB115 in this modification. Then, in step SB170 in the present modified example, the control unit 301 applies the click determined in step SA200 of FIG. 21 to the PC terminal 400 of a particular consumer who has not performed the click. Display information including advertisement information related to. Also in this modification, step SB170 corresponds to the recommended information output procedure described in each claim and functions as a recommended information output unit.

  In the modification described above, the control unit 201 of the calculation PC terminal 200 clicks on the My Page for the advertisement displayed on the PC terminal 400 of the specific consumer, in other words, the information related to the article. The UID of the specific consumer who performed is acquired. And the control part 201 accesses the advertisement database 1137 classified by user of the said DB server 100, and counts the number of the specific consumers from whom the said UID was acquired for every unit. Then, the control unit 201 determines advertisement information related to the click on the My Page to the consumer PC terminal 400 belonging to the unit according to the number of specific consumers counted for each unit. To do. As described above, in this modified example, not only the fact that the article is directly purchased, but also the fact that the click on the My Page, which is the previous stage leading to the purchase, is reflected in the advertising information. To decide. Thereby, the more effective recommendation which considered the potential consumption demand can be performed.

  In particular, the control unit 301 of the WEB server 300 relates to the click determined in the calculation PC terminal 200 for the PC terminal 400 of a consumer who has not performed the click among specific consumers. The display information of my page related to the specific consumer including the advertisement information is output. As described above, in this modification, the PC terminal 400 of a specific consumer who has not even performed the click on the advertisement on the My Page is related to the specific consumer including advertisement information related to the advertisement. Output the display information of My Page. This makes it possible to make an effective recommendation targeting a specific consumer who is highly likely not to be aware of the article at that time.

  Note that, in the above, the control unit 301 of the WEB server 300 acquires the advertisement information determined by the control unit 201 of the calculation PC terminal 200 from the user-specific advertisement database 1137 and includes the acquired advertisement information. Although the display information of the My Page related to the consumer is output to the PC terminal 400 of the specific consumer, the present invention is not limited to this. That is, the control unit 301 of the WEB server 300 acquires the determined advertisement information from the user-specific advertisement database 1137, and sends an email including the acquired advertisement information to the PC terminal 400 of at least one consumer. You may make it output.

  In the above, the advertisement database for each user 1137 is constructed by automatically storing the advertisement information determined by the control unit 201 of the calculation PC terminal 200. However, the present invention is not limited to this, and the user-specific advertisement database 1137 may be constructed by manually inputting and storing the advertisement information determined by the operator referring to the purchase history information or the click history information, for example. Good.

  Further, the system configuration of the product recommendation system 1 described above is an example, and the product recommendation system may be realized with other system configurations. Furthermore, each process such as a process related to clustering and a process related to determination of advertisement information for the consumer's PC terminal 400 may be executed by an apparatus different from the above-described embodiment and the modified example of (1).

  In addition, in the above, the arrow shown in FIG. 1 shows an example of the signal flow, and does not limit the signal flow direction.

  In addition, the flowcharts shown in FIGS. 12, 13, 16 and the like do not limit the present invention to the procedure shown in the above-described flow, and the addition / deletion or order of the procedures within the scope of the gist and technical idea of the invention. May be changed.

  In addition to those already described above, the methods according to the above-described embodiments and modifications may be used in appropriate combination.

  In addition, although not illustrated one by one, the present invention is implemented with various modifications within a range not departing from the gist thereof.

1 Product recommendation system 100 DB server 200 PC terminal 200 for calculation
201 control unit 300 WEB server 301 control unit 400 PC terminal (information terminal)
420 display 421 keyboard 422 mouse 1132 singing history database (first consumer database)
1133 User profile database (clustering database)
1134 Purchase history database (second consumer database)
1137 Advertising Database by User 1139 Authentication Database 1140 Click History Database (Second Consumer Database)

Claims (7)

A first consumer database storing consumption history information in which identification information of a plurality of first products related to the first business and consumption behavior histories of a plurality of consumers for the first product are associated with each first product. A history acquisition procedure for accessing and acquiring consumption history information of a plurality of consumers for a plurality of first products;
Using the consumption history information of the plurality of consumers for the plurality of first products acquired in the history acquisition procedure, generating clustering information in which each consumer's identification information is associated with one of a plurality of units A clustering procedure to
Consumption history information in which identification information of a plurality of second products related to the second business and consumption behavior histories of a plurality of consumers or specific consumption-related behavior histories other than the consumption for the second product are associated with each second product. Identification information acquisition that accesses the second consumer database stored for each and obtains identification information of each of a plurality of consumers who have performed consumption behavior or specific consumption-related behavior other than consumption for a predetermined second product Procedure and
A consumer counting procedure for accessing a clustering database storing the clustering information generated in the clustering procedure and counting the number of consumers whose identification information has been acquired in the identification information acquisition procedure for each unit;
According to the number of consumers counted for each unit in the consumer counting procedure, the consumption behavior for the information terminal of at least one consumer among the plurality of consumers associated with the unit or the A recommended information determination procedure for determining recommended information related at least to consumption-related behaviors;
A recommended information output procedure for obtaining recommended information determined in the recommended information determination procedure from a recommendation database, and outputting the acquired recommended information to the information terminal of the at least one consumer;
A product recommendation method characterized by comprising: The recommended information determination procedure includes:
For each unit, the recommended information related to the second product having a large number of consumers who have performed the consumption behavior or the specific consumption-related behavior is obtained from the consumer who has performed the consumption behavior or the specific consumption-related behavior. The product recommendation method according to claim 1, wherein the determination is performed so that priority is given to the recommendation information related to the second product with a small number. The recommended information output procedure is as follows:
Consumers who are associated with the recommended information determined in the recommended information determination procedure for each unit and who have not performed the consumption behavior or the specific consumption-related behavior for the second product related to the recommendation information The product recommendation method according to claim 1, wherein the product is output to the information terminal. An authentication request receiving procedure for receiving an authentication request signal from the information terminal of a specific consumer;
Based on the authentication request signal received in the authentication request receiving procedure, the authentication database storing the authentication information of the specific consumer in advance for the first business is accessed, and the specific consumer related to the authentication request signal is accessed. An authentication procedure for performing authentication of
The recommended information output procedure is as follows:
Display information for the information terminal of the specific consumer, including the recommended information determined in the recommended information determination procedure for the specific consumer authenticated in the authentication procedure, to the information terminal of the specific consumer The product recommendation method according to any one of claims 1 to 3, wherein the product recommendation method is output. The identification information acquisition procedure includes:
As the specific consumption-related behavior, a specific operation is performed on the information related to the second product displayed on the information terminal of the specific consumer on the information screen displaying the display information. To get consumer identity information for
The consumer counting procedure includes:
Accessing the clustering database, counting the number of specific consumers whose identification information was acquired in the identification information acquisition procedure for each unit;
The recommended information determination procedure includes:
According to the number of the specific consumers counted for each unit in the consumer counting procedure, the predetermined information on the information screen is sent to the information terminal of the at least one consumer associated with the unit. The product recommendation method according to claim 4, wherein the recommendation information related to an operation is determined. The recommended information output procedure is as follows:
The display information including the recommended information related to the predetermined operation determined in the recommended information determination procedure for the information terminal of the consumer who has not performed the predetermined operation among the specific consumers The product recommendation method according to claim 5, wherein: A first consumer database that stores, for each first product, consumption history information in which identification information of a plurality of first products related to the first business is associated with consumption behavior histories of a plurality of consumers for the first product; ,
History acquisition means for accessing the first consumer database and acquiring consumption history information of a plurality of consumers for a plurality of first products;
Using the consumption history information of the plurality of consumers for the plurality of first products acquired by the history acquisition means, generating clustering information in which each consumer's identification information is associated with one of a plurality of units Clustering means to
Consumption history information in which identification information of a plurality of second products related to the second business and consumption behavior histories of a plurality of consumers or specific consumption-related behavior histories other than the consumption for the second product are associated with each second product. A second consumer database stored for each
Identification information acquisition means for accessing the second consumer database and acquiring identification information of each of a plurality of consumers who have performed a specific consumption-related behavior other than consumption behavior or consumption for a predetermined second product;
A clustering database for storing the clustering information generated by the clustering means;
Consumer counting means for accessing the clustering database and counting the number of consumers whose identification information has been acquired by the identification information acquiring means for each unit;
Depending on the number of consumers counted for each unit by the consumer counting means, the consumption behavior or at least one consumer information terminal among a plurality of consumers associated with the unit A recommended information determination means for determining recommended information related at least to consumption-related behavior;
Recommended information output means for obtaining recommended information determined by the recommended information determining means from a recommendation database, and outputting the acquired recommended information to the information terminal of at least one consumer;
A product recommendation system characterized by comprising:

Images