JP2011221608A - Recommendation information generation apparatus, program and terminal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide recommendation contents based on a recommendation method whose property is required to be known while considering which recommendation method out of a plurality of sorts of recommendation methods is suitable for a user.SOLUTION: A user terminal 100 and a server 400 including a recommendation information generating function are connected to a network 300. A processing part 410 in the server 400 comprises: recommendation order generation means M1 for generating a recommendation contents list composed of recommendation contents respectively obtained by a plurality of recommendation methods and the recommendation order of the recommendation contents; and selected recommendation method acquisition means M3 for acquiring a recommendation content selected by the user terminal 100 and an identifier of a recommendation method from the plurality of recommendation content lists generated by the means M1. The processing part 410 forms a database D1 for storing a group of the recommendation content selected by a user and the identification of the recommendation method and a database for storing the recommendation order of the recommendation contents of the recommendation method in a storage part 420 and generates recommendation information.

Description

  The present invention relates to an information recommendation presentation system, and more particularly to a recommendation information generation technique for generating recommendation information according to a usage history of a content user.

  Conventionally, various recommendation methods and recommendation engines have been used in a system that selects objects, information, or products that are considered useful for a user and recommends them according to the purpose of the user.

  A collaborative filtering method (Non-Patent Document 1) that recommends a similar user based on a user's behavior history and another user's behavior history, and recommends content that the similar user often browses, and content as a feature vector in the feature space. There is a content-based filtering method (Non-Patent Document 2) that recommends similar unselected or unevaluated content that is expressed and close in distance to the content selected by the user or the evaluated content.

  The recommendation method usually has different reasons and characteristics for recommendation, and the recommendation order is assigned to the content by the recommendation method.

  Conventional recommendation presentation systems use one or more of these methods to recommend content.

G. Linden et al .: “Amazon.com Recommendations, Item-to-Item Collaborative Filtering”, IEEE Internet Computing, 2003. Frontier of statistical science "Statistics of language and psychology", Iwanami Shoten, p202

  Some users are concerned about the purchase status of others, others are interested in something similar to their purchase status, and the recommended recommendation method varies depending on what they are purchasing. However, at present, the recommendation information provider selects the recommendation method uniformly for the user and provides the recommended content, so it is not possible to obtain the recommended content with the nature that you want to know. There are many cases.

  An object of the present invention is to provide a recommendation information generation device, a program, and a terminal that can provide recommended content by an appropriate recommendation method according to an individual and needs while avoiding the above problems.

  In order to achieve the above object, in the present invention, a recommendation information generating apparatus for providing recommendation information for recommending content to a user, comprising a processing unit and a storage unit connected to each other, the processing unit Includes a recommendation content generation unit that generates a recommended content list including recommended content and recommendation rankings according to each of the plurality of recommendation methods, and a recommended content selected by the user from the plurality of recommended content lists generated by the recommendation order generation unit. A recommendation information generating apparatus having a selection recommendation method acquisition unit for acquiring an identifier of a recommendation method, and a recommendation information generation program thereof are provided.

  In order to achieve the above object, according to the present invention, an administrator terminal of a system that uses a recommendation information generation device that provides recommendation information for recommending content to a user, and includes a display unit, On the display unit, the recommended information generation device generates a recommended content list including recommended content and a recommendation order of each of the plurality of recommended methods, and presents it to the user. As a result, the recommended content selected by the user and the identifier of the recommended method are acquired. An administrator terminal configured to display the usage rate of the recommendation method calculated in the above manner is provided.

  That is, in order to solve the above-described problem, in a preferred aspect of the present invention, a server including a processor that performs arithmetic processing, a storage unit connected to the processor, and an interface connected to the processor is provided. The processor is configured to generate a recommended content and a recommendation order according to a plurality of recommendation methods, a display control unit that controls to display a plurality of types of recommended content lists, and a user to select from the displayed recommended content list An acquisition means for acquiring a set of identifiers of recommended methods for generating the recommended content and recommended content, and a storage unit storing a database storing a set of identifiers of the recommended method for generating the content selected by the user and the recommended content Recommendation information with a database that stores content recommendation rankings And generating device.

  That is, in a preferred aspect of the present invention, a server that is a recommendation information generation device presents content recommendations to the user for each recommendation method, and the user selects a content of interest from the set of recommended content and the user has Can be held as a set of content. The identifier of the content selected by the user, the identifier of the content recommendation method, and the identifier of the user are transmitted to the server, and the server stores these pieces of information in the database. Based on the stored information, the recommendation method priority calculating means calculates the priority of the recommendation method. The server switches and outputs a recommendation method to be displayed to the user and a content set generated thereby according to the calculated priority of each recommendation method.

  Further, according to the present invention, the priority is calculated by performing a weighting process according to the selection frequency of the recommendation method that recommended the content selected by the user, and the priority is calculated according to the result.

  ADVANTAGE OF THE INVENTION According to this invention, the recommendation system shown to a user can be switched and displayed according to the selection condition of the recommendation system which recommended the content which the user selected, a user's preference with respect to a recommendation reason can be determined, and various recommendation systems Can be displayed in the order of recommendation methods that the user is interested in.

It is a block diagram which shows the structure of the recommendation information presentation system of a 1st Example. It is a sequence diagram which shows the whole process of the system based on a 1st Example. It is a flowchart which shows the recommendation system priority calculation means of a server based on a 1st Example. It is a flowchart which shows the priority calculation method of the recommendation system of a server based on 1st Example. It is a flowchart which shows the priority calculation method using the selection frequency of the recommendation system of a server based on a 1st Example. It is a flowchart which shows the recommendation method priority supplement method of the server based on 1st Example. It is a figure which shows the recommendation information display screen example of a user terminal based on a 1st Example. It is a figure which shows the recommendation information display screen example of a user terminal based on a 1st Example. It is explanatory drawing of the user selection content database based on a 1st Example. It is explanatory drawing of the content recommendation ranking database based on a 1st Example. It is explanatory drawing of the recommendation system priority database based on a 1st Example. It is a flowchart which shows the recommendation method utilization factor calculation means of a server based on 2nd Example. It is a figure which shows the recommendation system utilization factor display screen example of an administrator terminal based on a 2nd Example. It is a figure which shows the example of a recommendation information display screen which can register the content hold | maintained of a user terminal according to a 1st Example according to a use. It is a figure which shows an example of the table | surface which shows the weight of a display area based on 1st Example.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In this specification, in addition to “program”, terms such as “function”, “means”, and “unit” are used as expressions of various software executed by a processing unit such as a processor in a computer or various terminals. May be used. For example, the “recommendation order generation program” is referred to as “recommendation order generation function”, “recommendation order generation means”, and “recommendation order generation unit”. It should be noted that a recommendation engine that recommends content may be referred to as a “recommendation method” or “recommendation means”.

  The recommendation information presentation system of the first embodiment shown in FIG. 1 includes a user terminal 100, an administrator terminal 200, and a server 400. The administrator terminal 200 is not essential, but may be provided in the usage form. The user terminal 100, the administrator terminal 200, and the server 400 are connected via the network 300.

  The terminal devices such as the user terminal 100 and the administrator terminal 200 are computers each including a processor (not shown), a memory and storage constituting a storage unit, a communication interface, and a user interface. In FIG. 1, reference numerals 101 and 201 schematically show display units constituting the user interface. Note that these terminal devices may be mobile terminals such as mobile phones and PDAs that run application programs (for example, web browsers). Needless to say, each includes a display unit as a user interface, and display data from the server 400 received through the communication interface can be displayed using its own display control unit.

  Similarly, the server 400 is a computer including a processor that is the processing unit 410, a memory and a storage device that configure the storage unit 420, and a communication interface that is not illustrated. A processor such as a central processing unit (CPU) which is the processing unit 410 of the server 400 includes a recommendation rank generation unit M1 that generates a recommended content and a recommendation rank according to a recommendation unit and a recommendation method to be used. Display control means M2, selection recommendation method acquisition means M3, and recommendation method priority calculation means M4 are provided. Each of these functions is provided by a program stored in the memory. The program itself is stored in a storage device that constitutes the storage unit 420. The storage device 420 of the server 400 also stores a user-selected content database D1, a content recommendation ranking database D2, and a recommendation method priority database D3. Note that the recommendation method usage rate calculation means M5 is used in the second embodiment and will not be described in this embodiment.

  Next, the overall flow of the recommendation information generation system of this embodiment will be described with reference to the sequence diagram of FIG. In the figure, first, the server 400 obtains information about the content selected and held as content of interest to the user of the user terminal 100 from the user selection content database D1 (S101), and the recommendation means priority calculation means M4 determines the recommendation means. A priority is acquired (S102). The screen displayed on the user terminal 100 is switched and output according to the obtained priority of the recommendation means (S103). The retained content group and the content group based on the recommendation method are displayed on the display screen of the user terminal 100 on the client side (S104).

  The user performs an operation S105 in which the user terminal 100 registers content of interest from the recommended content group displayed on the display screen of the user terminal 100 as content to be held. In response, the selected content, the recommended method of the selected content, and the user identifier are transmitted from the client side to the server (S106). The server 400 side registers the transmitted selected content, the recommended method of the selected content, and the user identifier in the user selected content database D1 of the database (S107, S108). Here, the operations S105 to S108 for registering the content that the user is interested in as the retained content may be repeated.

  Next, the operation of the recommendation method priority calculation means M4 in the server 400 of the system of the present embodiment will be described using the flowchart of FIG. In the figure, the server 400 acquires data that matches the user identifier from the user-selected content database D1 (M401), and ends if there is no data that matches the user identifier. Then, the server 400 calculates the priority of the recommendation method by the number of recommendation engines, that is, the number of recommendation methods, in operations M402 to M405. Here, several methods can be considered for calculating the priority in the operation 404. In the present embodiment, two types of priority calculation methods for the recommendation method for calculating the priority will be described with reference to the flowcharts of FIGS. 4 and 5.

  First, an example of a priority calculation method and recommendation method priority calculation means in this embodiment will be described with reference to the flowchart of FIG. User-selected content that matches the recommended method for calculating the user identifier (ID) and priority is acquired from the user-selected content DBD1 stored in the storage unit 420 (M4101). The order of the content in the content recommendation method is acquired (M4104). The reciprocal of the content order is added to the priority of the recommendation method (M4105). With regard to the recommendation method for which priority is to be calculated, all contents selected by the user are processed (M4102, M4103, M1406). The obtained result is set as the priority of the recommendation method. By calculating the priority, it is possible to increase the priority of the recommendation method in which the content with the recommendation method ranked higher is selected, and the order given by the recommendation method is closer to the user's preference. The higher the priority can be set.

  Next, another example of the priority calculation method and recommendation method priority calculation means in this embodiment will be described with reference to the flowchart of FIG. First, user-selected content that matches a user's identifier (ID) and recommendation method for which priority is to be calculated is acquired from the user-selected content database D1 (M4201). Next, when the method for recommending the content selected by the user is a recommendation method for which priority is to be calculated, a score, for example, 1 is added to the priority (M4204). Processes M4202, M4203, and M4205 are performed for all contents selected by the user, and the priority of the recommendation method is obtained. This process makes it possible to obtain the frequency with which the recommendation method is selected by the user, and it is possible to set a higher priority for a recommendation method that is easier to select.

  Depending on the WEB site that employs many recommendation methods, there may be a recommendation method that has not been touched by the user by changing the priority of the recommendation method and switching the display. It is also possible that there was no opportunity to select content from. In such a case, the recommendation method priority of the user can be set using the tendency of the other user.

  A recommendation method priority calculation method and recommendation method priority calculation means using the tendencies of other users will be described with reference to the flowchart of FIG. First, a priority is calculated for a recommendation method for which content has already been selected for each user. As the calculation method, the priority calculation method described in FIGS. 4 and 5 can be used. These priorities are registered in the recommendation method priority database D3. The priority of the registered recommended method of the user is acquired from the recommended method priority database D3 (M4301). For the recommended method whose priority is already registered, the recommended method similarity with other users is calculated (M4302 to M4305). Another user with a high recommendation method similarity is obtained and the priority of the recommended user is set as the priority of the unregistered method (M4306). In other words, by complementing the recommendation engine that is not registered by the user by using the priority of the recommendation engine of the other user who has the closest similarity, it is possible to use the tendency of content selection of the other user.

  FIG. 7 shows an example of a recommended information display screen displayed on the user terminal 100 based on the display data generated by the recommended content display control means M2 of this embodiment. In this example, three types of recommendation methods are used, the purchase ranking of the content is calculated, and the top few cases of purchase ranking are displayed. Content recommended by a method of displaying content purchased by a user or a content-based filtering method of displaying other content similar to that purchased by the user is displayed in display areas 702, 703, and 704, respectively.

  The user can select content that meets his needs from these contents and press an add to my content button 706 to register the content in an area 701 that holds interesting content that the user intends to purchase from now on. . Note that a selection check box is displayed for each content in each display area. In the figure, reference numeral 705 denotes a fixed pin mark, and the purchase ranking display area 702 indicates a state where the display position in the recommendation information display screen 700 is fixed by the fixed pin mark 705.

  In such a recommendation information display screen, the recommendation method to be displayed is changed or the display order is changed according to the priority of the recommendation means. Here, like the content 4, the same content with different recommendation methods may be displayed with the display order changed.

  FIG. 8 shows another example of the recommended information display screen displayed on the user terminal 100 under the control of the recommended content display control means M2 in this embodiment. As shown in the figure, the content information for each recommendation method may be displayed by tabs 802, 803, and 804. In addition, the number of content displays may be displayed as many as the display screen allows. In the example, content that cannot be displayed at a time can be displayed by a scroll bar 805 or the like. Reference numeral 806 denotes an add button as in FIG.

  According to this example, when referring to different recommendation methods, it is necessary to switch the tabs 802, 803, and 804 to display the content. In addition, although content with a higher recommendation order is displayed at the time of initial display, in order to browse to lower content, it is necessary to use and display by scrolling using the scroll bar 805. Here, even if the final content is not selected, switching the tab to display another display screen or scrolling to display content that has not been displayed means that the user May want to know recommendations from that point of view. Therefore, when the final recommended content is not selected, the switching operation itself of the tabs 802, 803, and 804 and the moving operation itself with respect to the scroll bar 805 are recorded and reflected in the priority of the recommendation method. May be. For example, in the recommended information display screen 800 of FIG. 8, the server 400 processes the identifier of the recommended method corresponding to the newly selected tab and the information of the identifier of the recommended content newly displayed by using the scroll bar 805. It can be included in the acquisition information of the selection recommendation method acquisition means M3 of the unit 410, and this additional acquisition information can be used by the recommendation method priority calculation means M4 and reflected in the recommendation method priority.

  FIG. 14 shows another recommendation method display screen 707 obtained by partially changing the recommendation method display screen 700 of FIG. An example of a recommendation information display screen in a case where content that the user is interested in can be stored in the microcomputer content display area 708 of the recommendation method display screen 707 separately for different purposes is shown. In the example, the contents of interest are divided into categories for use such as “child”, “interior”, and “pet”, and tabs 709, 710, and 711 are selected and displayed.

  The priority of the recommendation method can be calculated according to the priority of the recommendation method for each category of the retained content. When the user changes the category to be displayed by selecting the tabs 709, 710, and 711, the display can be changed according to the priority of the recommendation method related to the category. This makes it possible to selectively display an appropriate recommendation method even when the recommendation method required by the user differs depending on the application.

  The contents of the database accumulated in the storage unit 420 of this embodiment will be described with reference to FIGS. First, FIG. 9 shows an example of the user-selected content database D1. The database D1 is a database that stores registration information when the user selects content recommended for the user and registers the content as his own content in, for example, the My Content folder.

  The user-selected content database D1 includes a user identifier (ID) D101, a content identifier (ID) D102, a recommendation method identifier (ID) D103, a ranking D104 according to a content recommendation method, a selection flag D105, a selection date D106, a display area D107 and category D108.

  For example, in the case of the first row of the database D10, the user 1 (U1) selects the content 1 (C1) from the purchase ranking (recommendation method E1) as shown in the display screen of FIG. An example of database registration is shown.

  The recommendation order of the content 1 in the recommendation method can be referred to from the content recommendation order database D2, and can be registered in the order D104 by the content recommendation means at the time of registration.

  Like the content 3 on the display screen of FIG. 7, it is also conceivable that the same content is recommended in a different order in a plurality of recommendation methods and recommendation means. When such content is selected as the content held by the user, only the content 3 information may be stored in the user-selected content database D1, or as the original recommendation means selected as in the example of FIG. The selection flag D105 is set to “1” so that it can be understood from which selection means the content has been selected. For the recommendation means including the same content, for example, the recommendation means E1, the selection flag D105 is set to “0”. As for the means E2, by setting the selection flag D105 to “1”, it is possible to explicitly indicate which recommendation means the user has used. Although the same content is displayed by other recommendation means, it is also conceivable that a weight is given to the recommendation means selected and reflected in the priority of the recommendation means.

  In addition, depending on the area where the content by the recommendation means is displayed, there is an area that is easily noticeable and easy to select. Therefore, it can be considered that the preference for the display area is related to the preference for the recommendation means. For example, on the display screen 700 of the user terminal as shown in FIG. 7, the purchase ranking located on the left is most noticeable. In a portable terminal, the higher the position, the more visible. For this reason, information indicating where the recommendation means is displayed is stored in the display area D107, and a weight corresponding to the visibility is given to each display area D107 and reflected in the priority of the recommendation means. Conceivable. For example, in FIG. 7, as is clear from the display area D107, “Purchase Ranking” is given a display area number 1, and “There is such a thing” is given a display area number 3.

  FIG. 15 is an example of a table in which the weight D402 representing the visibility is displayed according to the display area D401 in the present embodiment. The visibility by the display area varies depending on the display method such as the parallel display as shown in FIG. 7, the tab display as shown in FIG. 8, and the vertical scroll display on the mobile phone screen although not shown. It is preferable to prepare a table and store it in the storage unit.

  When the weight is reflected in the priority of the recommendation means, the reciprocal of the weight is applied to the priority calculation, so that the contribution to the priority of the recommendation means when the selection is made from the easy-to-see recommendation means is relatively Lower. As a result, the priority of the recommended means selected simply because it is easy to be noticed becomes higher, and further prevents the user's preference for the recommended means from being difficult to be reflected as a result of the circulation. be able to. On the other hand, since the weight when deliberately selected from recommendation means that are not easily noticeable increases, the intentional selection of the user is quickly reflected in the priority.

  It is also conceivable to display a recommendation means having a high priority in an area where the weight of the display area is large. In the table of FIG. 15, the display areas may be sorted according to the size of the weights, and recommendation means with higher priorities may be associated in order from the display areas with the highest weights. As a result, it becomes possible to place a more suitable one in a place where it is easy to see, leading to a reduction in operation time and an improvement in ease of use for the user.

  The priority of the recommendation method can be calculated according to the priority of the recommendation method described above for each category of the retained content. When the category displayed by the user is changed, the display can be changed according to the priority of the recommendation method related to the category.

  In order to calculate the priority of the recommendation method for each category, as shown in FIG. 9, information on which category of the retained content is registered in the category D108, and the priority calculation belongs to the category. It is possible to do it by content.

  In the present embodiment, the display of the recommendation means is changed according to the user's preference. However, since the display area and the state of the recommendation means displayed may be fixed, the purchase ranking in FIG. It is also conceivable to provide a function of fixing the display content / display area like a fixing means represented by a pin mark 705 on the area 702. The same applies to the tab-type display as shown in FIG. 8, and it is conceivable to provide a function for the user to explicitly select the tab display position and the selection of the tab displayed first.

  FIG. 10 shows an example of the content recommendation ranking database D2 of FIG. This database D2 is a database for storing the order that each recommendation means gives to the content. The content recommendation ranking database D2 includes a content identifier D201, a recommendation method identifier D202, and a ranking D203.

  If the content recommendation rank does not change for each user, for example, in the case of recommendation rank information by recommendation means such as purchase ranking, the recommendation rank is periodically calculated and registered in the content recommendation rank database D2. In the case of a recommendation method in which the order of contents changes depending on the user, the recommendation order is calculated when the user accesses the site and does not necessarily need to be registered in the database.

  FIG. 11 shows an example of the recommendation method priority database D3. This database D3 is a database that stores the priority calculated by the priority calculation means of the recommendation method. The recommendation method priority database D3 includes a user identifier D301, a recommendation method identifier D302, and a priority D303. Since the priority of the recommendation method is calculated according to the usage history of the user, it can be calculated, for example, at the end of the user's access and stored in the recommendation method priority database D3.

  Subsequently, an administrator terminal of a recommendation information generation system including a recommendation method statistical tool will be described as a second embodiment with reference to FIGS. On the side of providing an EC (Electronic Commerce) site or the like, a plurality of types of recommendation methods may be introduced and used for the purpose of increasing purchase behavior. In particular, when introducing a recommendation method of another company, there is a need to know the tendency of the recommendation method to be used, such as a usage rate for each recommendation method. In this embodiment, an example of a recommendation method statistical tool on the administrator side of the recommendation information generation system will be described.

  With reference to the flowchart of FIG. 12, the recommendation method utilization rate calculation means M5 of this embodiment will be described. First, a period during which the recommendation method is to be analyzed is acquired using the GUI of the administrator terminal 200. And the recommendation method information selected within the obtained predetermined period is acquired from the user selection content database D1 (M501).

  The total number of recommended methods used is set as T (M502). Next, the number of uses included in the recommendation method information acquired for each recommendation method is calculated (M503 to M507). For example, the usage number S [j] of the recommendation method E [j] is set (M505). Next, the usage rate of the recommendation method is calculated as the recommended method usage number with respect to the total recommendation method usage number. The utilization rate R [j] of the recommendation method E [j] is calculated as S [j] / T (M506). The usage rate of the recommendation method is displayed on the administrator terminal.

  The calculation of the usage rate of the recommendation method may be performed for the content type. When the content is a product, the usage rate can be calculated for each type of content such as a book, CD, or home appliance. Similarly, the usage rate of the recommendation method may be calculated according to the type of user. Users can be classified according to information such as gender, age, and site usage history, and the usage rate can be calculated for each type.

  Furthermore, as one of the purposes of the recommendation method, there is a point of providing the user with content information that is not easily noticeable to the user. Therefore, the number of types of content selected by the user can also be used as an effect evaluation of the recommendation method. In particular, it is possible to evaluate how much the user's eyes can touch a product positioned at a lower position after excluding higher sales.

FIG. 13 shows a display screen 1300 of the recommendation means usage rate display screen of the recommendation means analysis tool displayed on the display screen 201 of the administrator terminal 200 of this embodiment. A pull-down menu 1301 can be used to specify the analysis period and display the usage rate of each recommendation means. Similarly, it is possible to specify the content type, the user type, etc., and display the usage rate of the recommendation means for each category.
This allows site administrators to know which recommendations are highly effective,
It can also be used for applications such as selection of recommendation means.

  The present invention relates to an information recommendation presentation system, and is particularly useful as a recommendation information generation technique for generating recommendation information in accordance with a usage history of a content user.

DESCRIPTION OF SYMBOLS 100 ... User terminal 200 ... Administrator terminal 300 ... Network 400 ... Server 410 ... Processing part 420 ... Memory | storage part M1 ... Recommendation order generation means M2 ... Recommended content display control means M3 ... Selection recommendation system acquisition means M4 ... Recommendation system priority calculation Means M5... Recommendation method utilization rate calculation means D1... User selected content database D2... Content recommendation ranking database D3.

Claims (20)

A recommendation information generation device that provides recommendation information for recommending content to a user,
A processing unit and a storage unit connected to each other;
The processor is
A recommendation order generating unit that generates a recommended content list including recommended content and a recommendation order of each of the plurality of recommendation methods;
A selection recommendation method acquisition unit for acquiring the recommended content selected by the user and an identifier of the recommendation method from the plurality of recommended content lists generated by the recommendation order generation unit;
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
Having a recommendation method priority calculation unit for calculating the priority of the plurality of recommendation methods based on the set of the obtained identifiers;
A recommendation information generating apparatus characterized by the above. The recommendation information generating device according to claim 2,
The recommendation method priority calculation unit of the processing unit is:
Calculating the priority of the recommendation method according to the recommended content selected by the user and the selection frequency of the recommendation method;
A recommendation information generating apparatus characterized by the above. The recommendation information generation device according to claim 3,
The recommendation method priority calculation unit of the processing unit is:
When calculating the priority of the recommendation method not selected by the user, the selection frequency of the propulsion method selected by another user having a similar tendency of the recommendation method is used.
A recommendation information generating apparatus characterized by the above. The recommendation information generating device according to claim 2,
The recommendation method priority calculation unit of the processing unit is:
The priority of the recommendation method is calculated by weighting according to the ranking in the recommendation method of the recommended content selected by the user.
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
A recommended content display control unit configured to control a display area for displaying the recommended content list on a display screen of a terminal used by a user so as to be divided into a plurality of parts;
The selection recommendation method acquisition unit further acquires an identifier of the display area of the recommended content selected by the user.
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 6,
The processor is
A recommendation method priority calculation unit that calculates a priority of the plurality of recommendation methods after weighting the display region, based on a set of the recommended content selected by the user, the recommendation method, and the identifier of the display region Having
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
A recommended content display control unit for generating display data for displaying the recommended content list divided into a plurality of regions on a display screen of a terminal used by a user;
The recommended content display control unit weights each display area, and controls to arrange and display the recommended content generated by the recommendation method having a high priority according to the weight of the display area.
A recommendation information generating apparatus characterized by the above. It is the recommendation information generation device according to claim 8, Comprising:
The recommended content display control unit of the processing unit is
Based on the instruction from the terminal, the display area and the display content thereof can be fixed so that the display of the recommendation method displayed on the display screen of the terminal is not changed.
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
A recommended method usage rate calculation unit that calculates a usage rate of the recommended method based on a set of the recommended content selected by the user and the identifier of the recommended method, which is displayed on a display unit of a terminal used by an administrator;
A recommendation information generating apparatus characterized by the above. It is the recommendation information generation device according to claim 10, Comprising:
The recommendation method usage rate calculation unit of the processing unit calculates the usage rate of the recommendation method according to an analysis period, a content type, or a user type.
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
A recommended content display control unit configured to control the display area of the terminal used by the user to divide and display the display area for displaying the recommended content list;
The selection recommendation method acquisition unit acquires an identifier of the display area of the recommended content, the recommendation method, or the recommended content, which is operated by the user on the display screen.
A recommendation information generating apparatus characterized by the above. The recommendation information generating device according to claim 12,
The processor is
Having a recommendation method priority calculation unit for calculating the priority of the plurality of recommendation methods based on the set of the obtained identifiers;
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
Forming a database that stores a set of the recommended content selected by the user and the identifier of the recommended method in the storage unit;
A recommendation information generating apparatus characterized by the above. The recommendation information generating apparatus according to claim 1,
The processor is
Forming in the storage unit a database for storing a recommendation order of the recommended content of the recommendation method;
A recommendation information generating apparatus characterized by the above. A recommendation information generating program executed by the processing unit of a recommendation information generating apparatus that includes a processing unit and a storage unit and provides recommendation information for recommending content to a user,
The processing unit is
Generate a recommended content list consisting of recommended content by each of a plurality of recommendation methods and their recommendation order,
Obtaining the recommended content selected by the user and the identifier of the recommended method from the plurality of recommended content lists presented to the user;
A recommendation information generation program characterized by being executed as described above. A recommendation information generation program according to claim 16,
The processing unit is
Forming a database that stores a set of the recommended content selected by the user and the identifier of the recommended method in the storage unit;
A recommendation information generation program characterized by being executed as described above. A recommendation information generation program according to claim 16,
Forming in the storage unit a database for storing a recommendation order of the recommended content of the recommendation method;
A recommendation information generation program characterized by being executed as described above. An administrator terminal of a system that uses a recommendation information generation device that provides recommendation information for recommending content to a user,
A display unit, and the display unit includes:
As a result of the recommendation information generation device generating a recommended content list including recommended content and recommendation rankings according to each of a plurality of recommendation methods and presenting it to the user, the recommended content selected by the user and the identifier of the recommendation method are acquired. Display the usage rate of the recommended method calculated by
An administrator terminal characterized by that. The administrator terminal according to claim 19,
The display unit displays the usage rate of the recommendation method according to an analysis period, a content type, or a user type on a display screen of the display unit.
An administrator terminal characterized by that.

Images