JP2011065217A - Recommend information generation device, recommend information generation method and recommend information generation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a recommend information generation device for improving recommend effects for a user. <P>SOLUTION: This recommend information generation device includes: extracting first recommend merchandise information related with merchandise or a service to be recommended to a user based on the browsing history information of the user (S11 to S14); preliminarily extracting second recommend merchandise information based on first recommend merchandise information (S7); storing second recommend merchandise information by associating the second recommend merchandise information with the first recommend merchandise information (S8), and generating the first recommend display information for making a user terminal display the first recommend merchandise information (S15); generating second recommend display information for making the user terminal display the second recommend merchandise information by referring to the merchandise information storage means (S16); and transmitting the first recommend display information and the second recommend display information (S17). <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a technical field of a recommendation information generation device, a recommendation information generation method, and a recommendation information generation program for recommending a product or service to a user using user browsing history information.

  Conventionally, in order to promote transactions in electronic commerce systems, recommend systems that provide products that are expected to meet the user's preference from a large amount of information existing on the Internet, and advertisements posted on websites When a viewer purchases an advertiser's product, an affiliate that gives a success fee to a website operator is used. For example, as a recommendation system, Japanese Patent Laid-Open No. 2004-228867 manages not only the purchase history of content purchased on the online storage by the user but also the operation history of content that can only be acquired after online purchase by the user database. The content recommendation and the sales method for grasping the user's preference and recommending and selling the optimum content for the user are disclosed.

JP 2004-13231 A

  However, in the above-described prior art, if the products are narrowed down and recommended, the possibility of deviating from the user's preference increases. In addition, when multiple recommended products are presented, there is a high possibility that the user will hit the product that the user currently desires, but the direct relationship of the presented product is often low, and the user selects The product's appeal effect tends to be diminished. Therefore, in order to further promote transactions in the electronic commerce system, it is desired to improve the recommendation effect for the user.

  Also, user preferences are uncertain and change over time. For this reason, when recommending products by narrowing down the details, the user may be far from the products that the user really wants. In that case, it is preferable to make a configuration such that after recommending a product as if it is shaded, it can be further narrowed down and recommended in response to a user operation.

  The present invention has been made in view of such problems, and an example of the problem is to provide a recommendation information generation device, a recommendation information generation method, and a recommendation information generation program that improve a recommendation effect for a user. With the goal.

  In order to solve the above-mentioned problem, the invention according to claim 1 is characterized in that, based on the browsing history information of the user, first recommendation extracting means for extracting first recommended product information related to a product or service recommended to the user, and the first Based on the recommended product information, second recommendation extraction means for extracting second recommended product information in advance, and product information storage for storing the second recommended product information by associating the second recommended product information with the first recommended product information. Means, first recommendation generating means for generating first recommendation display information for displaying the first recommended product information on a user terminal, and the second recommended product information with reference to the product information storage means. Second recommendation generation means for generating second recommendation display information for display on a terminal; Transmitting means for transmitting the recommendation display information and the second recommendation display information, characterized by comprising a.

  According to a second aspect of the present invention, in the recommendation information generating device according to the first aspect, the second recommendation generating unit displays the second recommended product information in a balloon area based on the terminal operation of the user. 2 recommendation display information is generated.

  According to a third aspect of the present invention, in the recommendation information generating device according to the first or second aspect, the second recommendation generating means sends the second recommended product information together with the first recommended product information in a balloon area. Second recommendation display information to be displayed is generated.

  According to a fourth aspect of the present invention, in the recommendation information generating apparatus according to any one of the first to third aspects, user information collecting means for collecting the browsing history information from the user terminal and the browsing history information A browsing history storage means for storing the browsing history information, wherein the user information collecting means collects the browsing history information stored in the user terminal, and the first recommendation extracting means is based on the collected browsing history information. The first recommended product information is extracted, and the second recommendation extraction unit extracts the second recommended product information based on the browsing history information stored in the browsing history storage unit.

  According to a fifth aspect of the present invention, in the recommendation information generating apparatus according to the fourth aspect, the user information collecting unit collects the browsing history information of different domains browsed from the user terminal in units of the user terminal. It is characterized by that.

  6. The recommendation information generating apparatus according to claim 5, wherein after browsing history information of different domains is stored in a cache, a first recommendation extraction unit refers to the cache and extracts first recommended product information. To do.

  The recommendation information generating apparatus according to any one of claims 1 to 6, wherein the second recommendation extracting unit extracts second recommended product information based on user information.

  The invention according to claim 8 is a first recommendation extraction step for extracting first recommended product information related to a product or service recommended to the user based on the browsing history information of the user, and a second based on the first recommended product information. A second recommendation extracting step for extracting recommended product information in advance; a product information storing step for storing the second recommended product information by associating the second recommended product information with the first recommended product information; and the first recommended product. A first recommendation generating step for generating first recommendation display information for displaying information on the user terminal, and a second for displaying the second recommended product information on the user terminal with reference to the product information storage means. A second recommendation generating step for generating recommendation display information; and the first recommendation. A transmission step of transmitting the de-display information and the second recommendation display information, and having a.

  The invention described in claim 9 is based on the first recommended product information, the first recommended product information for extracting the first recommended product information related to the product or service recommended to the user based on the browsing history information of the user, Second recommendation extraction means for preliminarily extracting second recommended product information, product information storage means for associating the second recommended product information with the first recommended product information, and storing the second recommended product information, the first recommended product First recommendation generating means for generating first recommendation display information for displaying information on the user terminal, and second recommendation for displaying the second recommended product information on the user terminal with reference to the product information storage means Second recommendation generating means for generating display information, and the first recommendation Characterized in that to function as a transmitting means for transmitting the shown information and the second recommendation display information.

  According to the present invention, the first recommended product information related to the product or service recommended to the user is extracted based on the browsing history information of the user, the second recommended product information is extracted in advance based on the first recommended product information, and the first recommended product information is extracted. The second recommended product information is associated with the product information, the second recommended product information is stored, the first recommendation is generated to generate the first recommended display information for displaying the first recommended product information on the user terminal, and the product information is stored. By generating the second recommended display information for displaying the second recommended product information on the user terminal with reference to the means, and transmitting the first recommended display information and the second recommended display information, the user is interested. The product group assumed to be present is presented to the user as the first recommended product information, and the first recommended By presenting the second recommended product information related to the product information from the storage device stored in advance according to the user's operation, the relevance of the product information is generated, the appeal of the product is enhanced, and the recommendation effect for the user is improved. In addition, a precise product recommendation can be provided quickly and with a reduced load on the server.

It is a mimetic diagram showing an example of outline composition of a recommendation system concerning an embodiment of the present invention. It is a block diagram which shows an example of schematic structure of the recommendation server of FIG. It is a block diagram which shows an example of schematic structure of the user terminal of FIG. It is a sequence diagram which shows the operation example of browsing history preservation | save in the service provision server of FIG. It is a sequence diagram which shows the operation example of the recommendation goods pre-generation in the recommendation server of FIG. It is a sequence diagram which shows the operation example of the display method of the recommendation goods in the recommendation server of FIG. It is a schematic diagram which shows an example of the 1st recommendation merchandise information displayed on the user terminal of FIG. It is a schematic diagram which shows an example of the 2nd recommendation merchandise information displayed on the user terminal of FIG.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  First, a schematic configuration and functions of a recommendation system according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic diagram illustrating a schematic configuration example of a recommendation system according to an embodiment of the present invention.

  As shown in FIG. 1, the recommendation system 1 provides a recommendation server 10 for recommending a product or service recommended to the user, a product or service, a Web information page related to the product or service, a product, or the like. Service providing servers 12 and 14 that perform the purchase processing of the user, and a user terminal 20 for a user to browse Web information.

  The recommendation server 10, the service providing servers 12 and 14, and the user terminal 20 are connected by a network 5 such as the Internet. The recommendation server 10 and the service providing servers 12 and 14 may be connected by a local area network. The user terminal 20 may be a mobile terminal connected through a mobile communication network (packet network).

  The recommendation server 10 includes user information such as user attributes and browsing history, first recommended product information related to products or services recommended to the user based on the user information, second recommended product information based on the first recommended product information, and the like. Is stored in the database 11. Then, the recommendation server 10 functions as an example of a recommendation information generation device, and extracts first recommended product information based on the user's browsing history information or extracts second recommended product information based on the first recommended product information. To do. Further, the recommendation server 10 generates a document on the Web described in a markup language such as HTML (HyperText Markup Language). For example, the recommendation server 10 generates first recommended display information for displaying the first recommended product information on the user terminal 20 or second recommended display information for displaying the second recommended product information on the user terminal 20. Or generate. Here, the user information includes user's browsing history, purchase history, behavior history such as tracking information on transition on the web, user personal information such as user ID and user's age and gender, and the like. The first recommended product information and the second recommended product information include product information and service information.

  The database 11 includes a hard disk drive, a silicon disk, and the like, and stores second recommended product information associated with the first recommended product information for each purchase history and user ID.

  The service providing servers 12 and 14 have databases 13 and 15, respectively. The service providing servers 12 and 14 provide various services such as book sales, airline ticket reservations, auctions, etc., function as Web servers, and perform various processes related to purchase procedures for products and the like. In the present embodiment, since the domains differ depending on various services such as book sales, airline ticket reservations, auctions, etc., the service providing server 12 and the service providing server 14 have different domain names.

  The databases 13 and 15 are composed of hard disk drives, silicon disks, and the like, and store product information for advertising products and services, user IDs, purchase histories, and the like.

Next, the configuration and function of the recommendation server 10 will be described with reference to the drawings.
FIG. 2 is a block diagram illustrating an example of a schematic configuration of the recommendation server 10.

  As shown in FIG. 2, the recommendation server 10 functioning as a computer includes a control unit 10a, a storage unit 10b, and a communication unit 10c. The control unit 10a, the storage unit 10b, and the communication unit 10c are connected by a system bus 10d.

  The control unit 10a has a CPU (Central Processing Unit) and the like, and is described in a markup language such as HTML as an example of generating first recommended product information and second recommended product information or generating recommended display information. Web documents are generated.

  The storage unit 10b includes a RAM (Random Access Memory), a ROM (Read Only Memory), and the like, stores an execution program and calculation results of the control unit 10a, and speeds up data reading as a cache function. To do. The storage unit 10 b is an example of a browsing history storage unit that stores browsing history information collected from the user terminal 20.

  The communication unit 10c is connected to the network 5 and controls the communication state with the service providing servers 12, 14 and the user terminal 20 or the like. The configuration of the service providing servers 12 and 14 is almost the same as that of the recommendation server 10.

Next, the configuration and function of the user terminal 20 will be described with reference to the drawings.
FIG. 3 is a block diagram illustrating an example of a schematic configuration of the user terminal 20.

  As illustrated in FIG. 3, the user terminal 20 that functions as a computer includes a display unit 21, an operation unit 22, a communication unit 23, a storage unit 24, and a control unit 25. The display unit 21 and the like are connected through the system bus 26.

  The display unit 22 is, for example, a CRT (Cathode Ray Tube) display, a liquid crystal display, or the like, and displays information such as characters and images.

  The operation unit 22 is, for example, a keyboard, a mouse, or the like, receives an operation instruction from an operator or the like, and outputs the instruction content to the system control unit 25 as an instruction signal.

  The communication unit 23 is connected to the network 5 and controls the communication state with the recommendation server 10 and the service providing servers 12 and 14.

  The storage unit 24 is, for example, a RAM, a ROM, a hard disk drive, a silicon disk, or the like, and is assigned with an area for storing cookie information, and stores a browsing history for each accessed domain.

  The control unit 25 has a CPU (Central Processing Unit) and stores browsing history information from the Web server in the cookie area of the storage unit 24 or displays recommended product information on the display unit 22 according to the recommendation display information on the Web browser. Or grasping the state of the operation unit 22.

  Next, the operation of the present embodiment will be described with reference to the drawings.

  FIG. 4 is a sequence diagram illustrating an operation example of browsing history storage in the service providing server 12 or the service providing server 14. Hereinafter, the service providing server 12 will be described, but the service providing server 14 operates in the same manner.

  First, the user uses the user terminal 20 to access the market of the service providing server 12 and perform a login operation. Then, the user performs an operation on the user terminal 20 to access the market of the service providing server 12 and browse the web page. A case where the user accesses the service providing server 12 will be described.

  Then, the user terminal 20 transmits a page request as shown in FIG. 4 (step S1). Specifically, the control unit 25 of the user terminal 20 transmits request information such as a URL (Uniform Resource Locator) to the service providing server 12.

  Next, the service providing server 12 transmits page data to the user terminal 20 (step S2). Specifically, in response to a page request from the user terminal 20, the service providing server 12 transmits Web page information described in a markup language such as HTML to the user terminal 20.

  Next, the service providing server 12 stores the browsing history (step S3). Specifically, the service providing server 12 stores the page request from the user terminal 20 in the database 13 as a browsing history. Thus, the service providing server 12 has a function as a user information collecting unit that collects user information from the user terminal 20, and also functions as an example of a recommendation information generating device. When the user purchases a certain product or the like, the service providing server 12 stores a purchase history.

  Next, the service providing server 12 generates cookie information for storing in the user terminal 20 regarding the browsing history (step S4). Specifically, the service providing server 12 specifies, as cookie information, an ID for identifying a browsing history such as a shop ID indicating which shop is browsed and an item ID indicating which product or service is browsed, Cookie information such as a service ID indicating whether the form is a mall or an auction is generated from the browsing history.

  Next, the service providing server 12 stores the browsing history in the cookie (step S5). Specifically, the service providing server 12 transmits the domain name related to the service providing server 12 as the domain attribute of the cookie together with the cookie information related to the browsing history. In the case of the service providing server 14, a different domain name is used.

  Next, the user terminal 20 stores the browsing history in a cookie (step S6). Specifically, the user terminal 20 stores the received cookie information in the storage unit 24 of the user terminal 20. The web browser stores about 70 to 80 browsing histories. When the upper limit is exceeded, the old browsing history is deleted.

Next, an operation example for generating in advance a product or service to be recommended to the user will be described with reference to the drawings.
FIG. 5 is a sequence diagram illustrating an operation example of the recommended product pre-generation in the recommendation server 10.

  First, the recommendation server 10 accesses the service providing servers 12 and 14, acquires products or services in the browsing history of each user, purchase history, user ID, and the like, and stores them in the database 11.

  Next, the recommendation server 10 extracts a second recommended product in advance based on the products and services of the browsing history and the user information (step S7). Specifically, the recommendation server 10 uses other user information together with user information such as purchase history, user age, and sex for information related to products and services in the browsing history as an example of the first recommended product information. Based on this, the second recommended product is extracted by collaborative filtering. As described above, the recommendation server 10 functions as an example of a second recommendation extraction unit that extracts the second recommended product information in advance based on the first recommended product information. Moreover, the recommendation server 10 functions as an example of a second recommendation extraction unit that extracts second recommended product information based on user information.

  Next, the recommendation server 10 stores the second recommended product information related to the second recommended product in the database 11 in association with the products and services in the browsing history and the user personal information (step S8). Specifically, the recommendation server 10 stores the second recommended product information in advance in the database 11 in association with the shop ID and item ID related to the products and services in the browsing history for each user ID. The cookie of the user terminal 20 has a browsing history (an example of first recommended product information) as simple information, and the recommendation server 10 receives the cookie information from the user terminal 20 and receives a detailed and detailed second recommendation. Make sure you can do it immediately. As described above, the database 11 functions as an example of a product information storage unit that stores the second recommended product information by associating the second recommended product information with the first recommended product information.

  Next, generation of recommendation display information in the recommendation server 10 and display of recommended products on the user terminal 20 will be described with reference to the drawings.

  FIG. 6 is a sequence diagram illustrating an operation example of a recommended product display method in the recommendation server 10. FIG. 7 is a schematic diagram illustrating an example of first recommended product information displayed on the user terminal 20. FIG. 8 is a schematic diagram illustrating an example of second recommended product information displayed on the user terminal 20.

  First, the user terminal 20 makes a page request to the recommendation server 10 (step S10). Specifically, the user uses the user terminal 20 to access a web page on which recommendation information is described. If the destination of the page request by the user terminal 20 is a Web page that makes a recommendation in the service providing servers 12 and 14, the user terminal 20 jumps to the recommendation server 10 by redirection or the like. Alternatively, the service providing servers 12 and 14 may request the recommendation server 10 for information on a portion for displaying the recommended product information, or may request generation of a Web page in which the recommendation information is described.

  Next, the recommendation server 10 requests cookie information from the user terminal 20 (step S11). Specifically, the recommendation server 10 requests browsing histories related to a plurality of domain names managed by the recommendation server 10 among the cookie information stored in the user terminal 20.

  And user terminal 20 transmits cookie information to recommendation server 10 (Step S12). Specifically, the user terminal 20 compares the requested domain names with the value of the domain attribute of the cookie, and collects the corresponding cookie information. And shop ID, item ID, etc. of the cookie information regarding the requested | required several domain name are transmitted to the recommendation server 10 with user ID. Note that the user ID may be an ID obtained by performing a login procedure on a website related to the recommendation server 10.

  Thus, the recommendation server 10 functions as an example of a user information collection unit that collects user information from the user terminal 20. The recommendation server 10 functions as an example of a user information collection unit that collects browsing history information stored in the user terminal 20.

  Next, the recommendation server 10 temporarily stores browsing history information of each domain in a cache (step S13). Specifically, the cookie information of each domain sent from the user terminal 20 is temporarily stored in the RAM of the storage unit 10b of the recommendation server 10. When domains are different for various services, browsing histories in all service domains are acquired by temporarily storing them in the cache in this way. As described above, the recommendation server 10 functions as an example of a user information collection unit that collects browsing history information, which is an example of user information, from the user terminal 20. Moreover, the recommendation server 10 functions as an example of a user information collection unit that collects browsing history information stored as cookie information in the storage unit 24 of the user terminal 20.

  Then, the recommendation server 10 collects the browsing history information of each domain (step S14). Specifically, the recommendation server 10 arranges the browsing history from the browsing history of each domain in order for each access time based on the information on the access time acquired from the cookie. In this way, browsing history information is collected for each user regardless of the domain name. This browsing history information is an example of first recommended product information. The browsing history information in which the browsing history information of each domain is collected for each user is stored in the cache area of the storage unit 10b as an example in the browsing history storage unit. The database 11 may be used instead of the storage unit 10b.

  Thus, the recommendation server 10 functions as an example of a first recommendation extraction unit that extracts first recommendation product information related to a product or service recommended to the user based on the user's browsing history information. Moreover, the recommendation server 10 functions as an example of a first recommendation extraction unit that extracts first recommendation product information based on the browsing history information. Further, the recommendation server 10 functions as an example of a first recommendation extraction unit that extracts first recommendation product information by referring to the cache after storing browsing history information of different domains in the cache. Further, the recommendation server 10 functions as an example of a user information collecting unit that collects browsing history information of different domains browsed from the user terminal 20 in units of user terminals.

  Next, the recommendation server 10 merges the top images of the products in the browsing history into the first recommendation pane (step S15). Specifically, the recommendation server 10 first extracts the first recommended product information based on the shop ID and item ID of the latest browsing history information. Then, the recommendation server 10 merges the products and services of the latest browsing history as the first recommended product information into the pane of the frame displaying the first recommended product information on the Web page that has received the request. . The web page in which the first recommended product information is merged is information described in a markup language or the like, and is an example of first recommendation display information. Thus, the recommendation server 10 functions as an example of a first recommendation generation unit that generates first recommendation display information for displaying first recommendation product information on the user terminal. Moreover, the recommendation server 10 functions as an example of a first recommendation extraction unit that extracts the first recommended product information based on the collected browsing history information.

  Next, the recommendation server 10 produces | generates the 2nd recommendation display information relevant to the goods of a high-order browsing history (step S16). Specifically, the recommendation server 10 uses the second recommendation described in a markup language or the like so as to form a device for displaying second recommendation product information related to the product of the higher browsing history on the Web page. Generate display information. More specifically, when the pointer of the operation unit 22 (mouse) of the user terminal 20 is hovered over the position where the first recommended product information is displayed in the pane of the frame displaying the first recommended product information. The second recommended product information is described in a markup language so that it is displayed in a display form such as a balloon. Thus, the recommendation server 10 functions as an example of a second recommendation generation unit that generates second recommendation display information for displaying the second recommended product information on the user terminal 20 with reference to the product information storage unit. Further, the recommendation server 10 functions as an example of a second recommendation generation unit that generates second recommendation display information for displaying the second recommended product information in the balloon area based on the user's terminal operation.

  Next, the recommendation server 10 transmits page information to the user terminal 20 (step S17). Specifically, the recommendation server 10 transmits information on the Web page including the first recommendation display information and the second recommendation display information to the user terminal 20. As shown in FIG. 7, in the user terminal 20, recently checked product images 31 a, 31 b, 31 c, 31 d, and 31 e are displayed as first recommended product information in the pane 31 of the Web page 30. As described above, the recommendation server 10 functions as an example of a transmission unit that transmits the first recommendation display information and the second recommendation display information.

  Next, the recommendation server 10 searches for the second recommended product information (step S18). Specifically, the recommendation server 10 searches the database 11 for the second recommended product information related to the product of the higher browsing history by the user ID and the item ID stored in the cache area of the storage unit 10b. Thus, the recommendation server 10 functions as an example of a second recommendation extraction unit that extracts the second recommended product information based on the first recommended product information. The recommendation server 10 functions as an example of a second recommendation extraction unit that extracts second recommended product information based on the browsing history information stored in the browsing history storage unit.

  Next, the user terminal 20 transmits trigger information to the recommendation server 10 (step S19). Specifically, as shown in FIG. 8, the user moves the mouse pointer 33 on the display unit 21 of the user terminal 20, and the pointer 33 moves to the area where the first recommended product information 31 b is displayed. When it is over, the user terminal 20 transmits trigger information for requesting the second recommended product information.

  Then, the recommendation server 10 receives the trigger information and transmits the second recommended product information to the user terminal 20 (step S20). Specifically, the recommendation server 10 transmits the second recommended product information related to the product of the upper browsing history. Then, as shown in FIG. 8, in the balloon (balloon) area 32, the second recommended product information 32a and 32b related to the first recommended product information 31b, the ranking 32c as the second recommended product information, and the like are displayed. As described above, the second recommended product information 32a, 32b, and 32c are displayed in the balloon region 32 together with the first recommended product information 31b. The recommendation server 10 functions as an example of a second recommendation generation unit that generates second recommendation display information for displaying the second recommended product information together with the first recommended product information.

  Note that the corresponding second recommended product information may be transmitted each time the display area of the other first recommended product information that has not been moused over yet is moused over. In this case, along with the trigger information, information indicating which position the first recommended product information is also transmitted. Further, the second recommended product information may also be transmitted when the page information is transmitted in step S7.

  As described above, according to the present embodiment, the first recommended product information 31a, 31b, 31c, 31d, 31e related to the product or service recommended to the user is extracted based on the user's browsing history information, and based on the first recommended product information. The second recommended product information 32a, 32b, 32c is extracted in advance, the second recommended product information is associated with the first recommended product information, the second recommended product information is stored, and the first recommended product information is displayed on the user terminal 20. Generating first recommendation display information for generating first recommendation display information, and generating second recommendation display information for displaying the second recommended product information on the user terminal 20 with reference to the product information storage means (database 11). By transmitting the first recommendation display information and the second recommendation display information, A storage device that presents a group of products assumed to be of interest to the user as first recommended product information, and pre-stores second recommended product information related to the first recommended product information in accordance with the user's operation By presenting the product information, the product information relevance is enhanced, the appeal of the product is enhanced, the recommendation effect on the user can be improved, and the detailed product recommendation can be made quickly and with reduced server load. Can provide. Moreover, the recommendation server 10 can be linked to conversion by improving the recommendation effect on the product or service.

  In addition, although the user's preference is uncertain and is likely to change with the passage of time, the recommendation server 10 receives the user's operation after recommending the product as if it is shaded using the first recommended product information. Since the second recommended product information can be further narrowed down, the product that the user really wants can be recommended.

  In addition, when the recommendation server 10 presents the second recommended product information after the first recommended product information, it is possible to narrow down the products step by step or to navigate the product or service. The power is increased and the recommendation effect for the user can be improved.

  Furthermore, the recommendation server 10 refines the products and services in detail by using the second recommended product information, and further adds explanations of the products and services, thereby increasing the appeal of the products and services and increasing the recommendation effect on the user. Can be improved.

  Also, the recommendation server 10 generates the second recommendation display information for displaying the second recommendation product information 32a, 32b, and 32c in the balloon area 32 based on the user's terminal operation as when the mouse pointer 33 is over the mouse. In this case, since the second recommended product information 32a, 32b, and 32c related to the first recommended product information 31b is accessed when the user accesses paying attention to the first recommended product information 31b or the like, the appeal power of the product or service is increased. As a result, it is possible to further improve the recommendation effect for the user, and it becomes possible to make an accurate product recommendation quickly and with a reduced load on the server. Further, since the user has selected the first recommended product information 31b, there is a high possibility that the current preference is indicated, and a second recommendation reflecting the current preference can be made. In addition, by the device for displaying the second recommended product information in the balloon area 32 based on the user's terminal operation, there is a possibility that the user may want to see other first recommended product information 31a, 31c, 31d, 31e as well as the first recommended product information 31b. Becomes higher.

  In addition, the recommendation server 10 displays the second recommended product information 32a, 32b, and 32c in the balloon region 32 together with the first recommended product information 31b so that the balloon region 32 is displayed from the first recommended product information 31b. When generating the recommendation display information, the relationship between the first recommended product information 31b and the second recommended product information 32a, 32b, and 32c is easily understood by the user, the appeal of the product or service is increased, and the recommendation effect for the user is further increased. Can be improved. In addition, it becomes easier for the user to grasp the flow of the recommendation, the appealing power of the product or service is increased, and the recommendation effect for the user can be further improved.

  Moreover, the recommendation server 10 collects browsing history information as simple information stored in the user terminal 20, and based on the browsing history information such as the collected shop ID and item ID, the first recommended product information and the second recommended product. When information is extracted, the first recommended product information 31a, 31b, 31c, 31d, 31e and the second recommended product information 32a, 32b, 32c can be quickly provided to the user. The recommendation effect with respect to can be further improved. Moreover, when extracting 2nd recommendation product information based on the browsing history information memorize | stored in RAM of the memory | storage part 10b of an example of a browsing history memory | storage means, an elaborate product recommendation is promptly performed in the state which reduced the load of the server. It becomes possible.

  Based on the browsing history information collected in advance by the service providing servers 12 and 14, the recommendation server 10 extracts the second recommended product information and associates it with the product or service shop ID, item ID, user ID, etc. of the browsing history. Since the second recommended product information is stored in the database 11, the first recommended product information and the second recommended product information can be quickly provided to the user by the shop ID, the item ID, the user ID, and the like. Thus, the recommendation server 10 and the service providing servers 12 and 14 are linked.

  In addition, when the recommendation server 10 collects browsing history information of different domains browsed from the user terminal 20 in units of user terminals based on the user ID or the like, the browsing history information that is different for each domain is unified for the user. Since the first recommended product information and the second recommended product information can be provided based on the information, the appeal of the product or service is enhanced, and the recommendation effect for the user can be further improved.

  In addition, when the recommendation server 10 previously extracts the second recommended product information based on the user information, the second recommended product information can be further refined, the appeal of the product or service is increased, and the recommendation effect for the user is further improved. be able to.

  Further, the recommendation server 10 can quickly provide the second recommended product information to the user by storing the first recommended product information and the second recommended product information in advance in the database 11 in association with each other.

  In addition, when the recommendation server 10 stores the browsing history information of different domains in the cache of the storage unit 10b and then extracts the first recommended product information with reference to the cache, the state is reduced quickly and the load on the server is reduced This makes it possible to make elaborate product recommendations.

  The first recommended product information may be based not only on browsing history information but also on other user information such as behavior history. In this case, the recommendation server 10 can provide more detailed first recommended product information. Further, the first recommended product information and the second recommended product information may be moving images in addition to text data and still images.

  Furthermore, the present invention is not limited to the above embodiments. Each of the embodiments described above is an exemplification, and any configuration that has substantially the same configuration as the technical idea described in the claims of the present invention and has the same operational effects can be used. It is included in the technical scope of the present invention.

1: Recommendation system 10: Recommendation server (recommendation information generation device)
10b: Storage unit (cache)
11: Database (product information storage means)
12, 14: Service providing server (recommendation information generating device)
20: User terminal

Claims (9)

First recommendation extraction means for extracting first recommended product information related to a product or service recommended to the user based on the user's browsing history information;
Second recommendation extraction means for extracting second recommended product information in advance based on the first recommended product information;
Product information storage means for storing the second recommended product information by associating the second recommended product information with the first recommended product information;
First recommendation generating means for generating first recommendation display information for displaying the first recommended product information on a user terminal;
Second recommendation generation means for generating second recommendation display information for displaying the second recommended product information on the user terminal with reference to the product information storage means;
Transmitting means for transmitting the first recommendation display information and the second recommendation display information;
A recommendation information generating apparatus comprising: In the recommendation information generating device according to claim 1,
The recommendation information generating apparatus, wherein the second recommendation generating means generates second recommendation display information for displaying the second recommended product information in a balloon area based on a terminal operation of the user. In the recommendation information generating device according to claim 1 or 2,
The second recommendation generating means generates second recommendation display information for displaying the second recommended product information in a balloon area together with the first recommended product information. In the recommendation information generating device according to any one of claims 1 to 3,
User information collecting means for collecting the browsing history information from the user terminal, and browsing history storage means for storing the browsing history information;
The user information collecting means collects the browsing history information stored in the user terminal;
The first recommendation extracting means extracts the first recommended product information based on the collected browsing history information,
The recommendation information generating apparatus, wherein the second recommendation extraction unit extracts the second recommendation product information based on browsing history information stored in the browsing history storage unit. In the recommendation information generating device according to claim 4,
The recommended information generation apparatus, wherein the user information collection unit collects the browsing history information of different domains browsed from the user terminal in units of the user terminal. In the recommendation information generating device according to claim 5,
A recommendation information generation apparatus, wherein after storing the browsing history information of different domains in a cache, a first recommendation extraction unit extracts first recommendation product information with reference to the cache. In the recommendation information generating device according to any one of claims 1 to 6,
The recommendation information generating apparatus, wherein the second recommendation extraction unit extracts second recommendation product information based on user information. A first recommendation extracting step of extracting first recommended product information related to a product or service recommended to the user based on the browsing history information of the user;
A second recommendation extracting step for preliminarily extracting second recommended product information based on the first recommended product information;
A merchandise information storage step for storing the second recommended product information in association with the second recommended product information to the first recommended product information;
A first recommendation generating step for generating first recommendation display information for displaying the first recommended product information on a user terminal;
A second recommendation generating step of generating second recommendation display information for displaying the second recommended product information on the user terminal with reference to the product information storage means;
Transmitting the first recommendation display information and the second recommendation display information; and
A recommendation information generation method characterized by comprising: Computer
First recommendation extraction means for extracting first recommended product information related to a product or service recommended to the user based on the user's browsing history information;
A second recommendation extracting means for extracting second recommended product information in advance based on the first recommended product information;
Product information storage means for storing the second recommended product information by associating the second recommended product information with the first recommended product information;
First recommendation generating means for generating first recommendation display information for displaying the first recommended product information on a user terminal;
Second recommendation generating means for generating second recommendation display information for displaying the second recommended product information on the user terminal with reference to the product information storage means; and
A recommendation information generation program that functions as a transmission unit that transmits the first recommendation display information and the second recommendation display information.

Images