JP6414192B2 - Information processing apparatus, information processing method, and information processing program - Google Patents

Description

The present invention relates to an information processing device, a terminal device, an information processing method, a display method, an information processing program, and a display program that perform information processing related to a display order determination method for an item set that satisfies a specific condition.

In recent years, digital contents such as documents, music, images, and videos distributed and provided using a network have been greatly increased due to advances in digital technology and network technology. Along with this, there is an increasing need for services that recommend items such as digital contents that meet user preferences. As a technique for meeting such needs, a technique for providing a recommended item using a use history such as a browsing history, an evaluation history, and a purchase history regarding a user's item has been developed.

For example, there is a recommendation process using collaborative filtering as one of the techniques for recommending items. In recommendation processing using collaborative filtering, a large number of usage histories are accumulated, and items that match the user's preference are recommended using the usage histories.

In addition, in order to maintain the user's interest in recommendation, a technique for changing a set of recommended items has been developed.
For example, Patent Literature 1 and Patent Literature 2 disclose a technique that preferentially recommends an unpresented item using a user's item access history and presentation status.

JP 2009-157500 A JP2008-146147

Patent Document 1 refers to a user's access history, excludes items that have been presented from items to be recommended, and recommends only items that have not been presented. Therefore, it is possible to always replace all recommended items.
Patent Document 2 discloses the number of times an item is presented by using a distribution counting sort for items that are sorted in advance according to an index such as popularity order or price order using the item presentation status for a keyword input by a user. By sorting in ascending order, it is difficult to present the previously presented item when another keyword is input, so the presented item can be changed.

However, in the above-mentioned Patent Document 1, every time the recommended item is browsed, all the recommended items are replaced. Therefore, the user says, “This recommendation changes every time. I felt "I don't know," and sometimes I couldn't get the reliability of the user's recommended items.
Patent Document 2 describes that it can be applied to an information recommendation device that recommends suitable information from data stored in a database. However, when a recommended item is presented to a user, a search is performed. Since there is nothing to correspond to the keyword, it is not clear whether the same effect as the system for inputting the search keyword can be obtained.

Then, this invention aims at promoting utilization of a user's service by presenting the recommended item which made the securing of the reliability with respect to a user's recommended item and maintenance of interest compatible.

The present invention is an information processing apparatus that provides item information to a user in order to solve the above-described problems of the related art, and is information that can identify an item related to item information that may have been browsed by the user. A browsing information storage unit that stores certain browsing specific information, and a recommendation candidate set that is a set of items for which the recommended value is calculated for each item and a recommended value that indicates the degree of recommendation for each item is calculated. Based on the browsing specific information corresponding to one user and the recommended value calculating unit that forms the browsing rate total value calculation for calculating the browsing rate total value indicating the degree of browsing of the item by the one user for each item For each item included in the recommendation candidate set, the higher the recommendation value is, the higher the priority is, using the recommended value and the browsing rate aggregation value. A display order determination unit that calculates a priority that lowers the priority as the browsing rate aggregate value is higher, and determines a display order of item information to be provided to the one user according to the priority. Provided is an information processing apparatus including the information processing apparatus.

According to the information processing device, the terminal device, the information processing method, the display method, the information processing program, and the display program of the present invention, it is possible to present a recommended item that ensures both the reliability of the user's recommended item and the maintenance of interest. Thus, the use of the user's service can be promoted.

It is a figure which shows the structure of the whole system in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the information processing server apparatus 1 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the user information storage part 131 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the item information storage part 132 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the utilization log | history storage part 133 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the recommendation item information storage part 134 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the display information storage part 135 in the 1st Embodiment of this invention. It is a figure which shows the example of the display information memorize | stored in the 2nd storage format of the display information storage part 135 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the display total information storage part 136 in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the 1st display information total process in the 1st Embodiment of this invention. It is a correspondence table of the final display order and viewing rate in the 1st embodiment of the present invention. It is a display example of the recommendation page currently displayed on the terminal display part 34 in the 1st Embodiment of this invention. It is a correspondence table of page information and browsing constants in the first embodiment of the present invention. It is a flowchart which shows the procedure of the display total registration update process of step S104 in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the 2nd display information total process in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the display order determination process in the 1st Embodiment of this invention. It is a flowchart which shows an example of a series of procedures which the terminal device 3 in the 1st Embodiment of this invention acquires recommendation item detailed information, and transmits a display history. It is a block diagram which shows the structure of the information processing server apparatus 4 in the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the information processing server apparatus 5 in the 3rd Embodiment of this invention. It is a figure which shows the storage state of the display information storage part 535 in the 3rd Embodiment of this invention. It is a flowchart which shows the procedure of the display order determination process in the 3rd Embodiment of this invention. It is a flowchart which shows the procedure of the 1st display information total process in the 3rd Embodiment of this invention. It is a figure which shows the example of a calculation process of the display probability estimated value in the 3rd Embodiment of this invention. It is a figure which shows the example of the calculation process of the expected value of the last display order in the 3rd Embodiment of this invention. It is a flowchart which shows the procedure of the 2nd display information total process in the 3rd Embodiment of this invention.

Hereinafter, an information processing device, a terminal device, an information processing method, a display method, an information processing program, and a display program according to the present invention will be described with reference to the accompanying drawings. The item in the present invention may be digital content such as text, voice, music, video, web page, and various articles, and may be information on financial products, real estate, and people. That is, the item in the present invention may be tangible or intangible, whether it is paid or free.

<First Embodiment>
Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a configuration diagram of the entire system according to the first embodiment of the present invention. In the system according to the present embodiment, the information processing server device 1 and one or more terminal devices 3 (3a to 3n) are connected via a network 2. In each embodiment, only the information processing server device 1 may function as an information processing device, or the information processing server device 1 may function as an information processing device in cooperation with the terminal device 3.

The network 2 is a network such as the Internet, for example, and mediates exchange of information between the information processing server device 1 and the terminal device 3.
The terminal device 3 is a general computer having a CPU, RAM, ROM, hard disk drive, network interface, or the like, or a portable information terminal such as a mobile phone, a smartphone, or a tablet PC. The terminal control unit 31 and the terminal communication unit 32 And terminal input unit 33
And a terminal display unit 34.
The terminal communication unit 32 is a unit for communicating with the information processing server device 1 via the network 2.

The terminal input unit 33 is configured so that, for example, the terminal device 3 is a PC (Personal Computer).
) Is an interface for the user to operate the terminal device 3 such as a mouse, a keyboard, or a mobile phone, such as a button.
The terminal display unit 34 is an interface for displaying various information and visually showing it to the user, such as a display.
The terminal control unit 31 performs usage history transmission processing, recommended item acquisition request transmission processing, item detailed information display processing, and display history transmission processing via the network 2.

The usage history transmission process is a process of transmitting a usage history to the information processing server device 1 when a user (user) who uses the terminal device 3 performs an operation of using some item. The usage history is information related to the use of the item of the user, and at least the user identifier that uniquely identifies the user, or the terminal identifier for uniquely identifying the terminal device 3, and the item to be used It contains the item identifier uniquely identified and usage date information indicating the usage date. In the present embodiment, a case where a user identifier is used will be described as an example, but the same applies to a case where a terminal identifier is used. The term user entity identifier is used as a general term for the meaning of the user identifier and terminal identifier combined. The usage time information may be information indicating only the date or may be information indicating in detail up to a millisecond unit, and may be information that can be understood up to a necessary unit according to the form of service.

The recommended item acquisition request transmission process is information when the user performs an operation to acquire recommended item detailed information or when the terminal device 3 performs an operation of automatically acquiring recommended item detailed information. This is a process of transmitting a recommended item acquisition request to the processing server device 1. The recommended item detailed information is information related to one or more items recommended to the user. The recommended item acquisition request includes at least the user identifier of the user. In addition,
An acquisition limit number for limiting the number of recommended items to be acquired may be associated with the recommended item acquisition request.
The item detailed information display processing is processing for receiving item detailed information from the information processing server device 1 and displaying the received item detailed information on the terminal display unit 34.

Item detailed information is information relating to one or more items. An example of the item detailed information is the recommended item detailed information described above. When a recommended item acquisition request is transmitted to the information processing server device 1, the recommended item detailed information corresponding to the transmitted recommended item acquisition request is received. It will be. The item detailed information may be a list of latest item information or information that is displayed to all users in a similar manner, such as a list of popular item information.

The received item detailed information does not need to be displayed collectively on one page, and may be displayed across a plurality of pages. For example, when the recommended item detailed information is displayed across a plurality of pages, first, a part of the recommended item detailed information including only information regarding the recommended items from the first to the fifth display order is received. It is displayed on the terminal display unit 34. Next, when the user performs an operation for acquiring information related to the next recommended item from the terminal input unit 33, a part of the recommended item detailed information including only information related to the recommended items from the display rank 6th to 20th May be received and displayed on the terminal display unit 34.

The display history transmission process is a process of transmitting a display history to the information processing server device 1. The display history is information for specifying an item displayed on the terminal display unit 34. Among the received item detailed information, the item identifier of the item displayed on the terminal display unit 34 of the terminal device 3 and the actual display are displayed. Including the combination of the final display order, which is the displayed order, the user identifier of the user, and the display time information indicating the display time. The display time information may be information indicating only the date, or may be information indicated in units of milliseconds, and may be information that can be understood up to a necessary unit depending on the form of service. When the item detail information is displayed across a plurality of pages, the final display order is given from the top to the order in which the items included in the page are displayed for each page.

Note that the recommended item detailed information is basically sorted in the order of the display order determined by the information processing server device 1, and when displayed on the terminal display unit 34, the display order is higher in the upper part of the page. Information about recommended items is displayed. In this case, the display order becomes the final display order as it is.
In addition, the recommended items included in the recommended item detailed information may not be displayed in the order of display order. For example, when randomly selecting and displaying from the recommended items included in the recommended item detailed information, or by determining the appearance probability in advance for each display order, and extracting the recommended item from the received recommended item detailed information according to the appearance probability There is a case. At this time, since the display order and the final display order are different, when displaying on the terminal display unit 34, the display order is also displayed in association with the recommended item to determine which item is recommended for the user. By giving the elements and switching each time they are displayed, you can have a sense of expectation about what will come out next.

The information processing server device 1 is a device that exchanges various information with the terminal device 3 via the network 2. The information processing server device 1 can be implemented as software (program) processing using a computer including a CPU, RAM, ROM, hard disk drive (HDD), network interface, and the like. In addition, you may comprise the information processing server apparatus 1 using a some computer. For example, to load balance
The distributed processing may be performed using a plurality of computers that perform processing corresponding to each unit of the information processing server device 1. Alternatively, the distributed processing may be performed in such a manner that the processing of a part of the information processing server device 1 is performed by a computer and the processing of another part is performed by another computer.
FIG. 2 is a configuration diagram of the information processing server apparatus 1 in the present embodiment. The information processing server device 1 in this embodiment includes an information processing server control unit 11, an information processing server communication unit 12, and an information processing server storage unit 13.

The information processing server communication unit 12 is a unit for communicating with the terminal device 3 via the network 2.
The information processing server storage unit 13 stores various data using a storage device such as a memory or an HDD. The information processing server storage unit 13 includes a user information storage unit 131, an item information storage unit 132, a usage history storage unit 133, a recommended item information storage unit 134, a display information storage unit 135, and a display total information storage unit 136. It consists of.
The user information storage unit 131 stores a plurality of user information. FIG. 3 shows a user information storage unit 1
FIG. The user information associates a user identifier (user_id) that uniquely identifies a user of the terminal device 3 and user attribute information (user_info), and is stored in a table format as shown in FIG. User attribute information includes the user's name, age, gender, address (region), hobby, and membership date (year, month, date, date, etc.)
Information such as e-mail address and telephone number. If information processing server device 1 can purchase a product, it may include credit card information for paying the product.

The item information storage unit 132 stores a plurality of item information. FIG. 4 is a diagram illustrating a storage state of the item information storage unit 132. Item information is an item identifier (item_
id) and item attribute information (item_info) are associated with each other and stored in a table format as shown in FIG. The item attribute information is, for example, information such as an item name, a creator, a production year, a genre, a price, and a user condition suitable for using the item.
The usage history storage unit 133 stores a plurality of usage histories. FIG. 5 is a diagram illustrating a storage state of the usage history storage unit 133. As shown in FIG. 5, the user identifier (use) constituting the usage history
r_id), an item identifier (item_id), and usage time information (dl_date) are associated with each other and stored in a table format. When other information is included in the usage history, the storage format may be changed in accordance with the usage history so that all the information included in the usage history can be stored.

The recommended item information storage unit 134 stores a plurality of recommended item information. FIG. 6 (FIG.
FIG. 6B is a diagram illustrating a storage state of recommended item information. Below, two types of storage formats of recommended item information are demonstrated using FIG. The recommended item information is information used to create the recommended item detailed information described above.
The first storage format of the recommended item information associates a user identifier (user_id), an item identifier (item_id), and a recommended value (score) obtained by quantifying the degree of recommendation of an item to the user with reference to FIG. ). The first storage format of recommended item information is suitable when only one set of recommended item information is used for each user.

The second storage format of the recommended item information includes a user identifier (user_id), an item identifier (item_id), and a set identifier (group_i) for identifying a recommended item set.
d) and the recommended value (score) are associated with each other and stored in a table format as shown in FIG. The second storage format of the recommended item information is, for example, that there are a plurality of recommended pages, and the recommended item detailed information is created using a set of different recommended item information for each recommended page. Therefore, it is suitable when a set of a plurality of recommended items is required. For example, there are three recommended pages, p1, p2, and p3, and recommended pages p1 and p1.
3 creates recommended item detailed information using only the recommended item information of the set identifier g1,
When the recommended page p2 creates the recommended item detailed information using only the recommended item information of the collective identifier g2, when the user u1 accesses the recommended page p1 or p3, the recommended item information corresponding to the collective identifier g1 of the user u1 is displayed. When the recommended item detailed information created using the set is displayed on the terminal display unit 34 of the terminal device 3 and the user u1 accesses the recommended page p2, the set of recommended item information corresponding to the set identifier g2 of the user u1 is used. The recommended item detailed information created in this way is displayed on the terminal display unit 34 of the terminal device 3.

The display information storage unit 135 stores a plurality of display information created based on the display history. A method of creating display information will be described later. FIG. 7 (FIGS. 7A to 7B) is a diagram showing a storage state of display information. Hereinafter, two types of display information storage formats will be described with reference to FIG.
The first storage format of the display information includes a user identifier (user_id) and detailed display information (d
isp_info) and display time information (disp_date) are associated with each other in FIG.
). The display detailed information may be in any format as long as all displayed items and their final display order can be identified. In the first storage format of display information, one display information is created for each display history.

The second storage format of the display information includes a user identifier (user_id) and an item identifier (
item_id), final display order (disp_rank), and display time information (disp)
_Date) in association with each other and stored in a table format as shown in FIG. In the second storage format of display information, display information is registered for the number of displayed items for one display history. Specifically, for example, the item i7 is displayed in the final display order 1st in the terminal display unit 34 of the terminal device 3 used by the user u3, the item i3 is displayed in the final display order 2nd, When i5 is displayed in the third highest display order and the display time information at that time is March 9, 2009, as shown in FIG. 8, the display information storage unit 135 stores three display information. Will be.
The above is the description of the display information storage format. The detailed display information included in the display information in the first storage format of the display information and the combination of the item identifier and the final display order included in the display information in the second storage format of the display information are the browsing specifying information in this embodiment. Thus, the display information storage unit 135 becomes the browsing information storage unit in the present embodiment.

If the display history includes other information such as page information that is information related to the page on which the item is displayed, the display information may be created in association with the information. Of the page information, information related to the recommended page is used as recommended page information. In addition, in the display information related to the recommended item detailed information, if a set identifier that specifies a set of recommended item information that is the basis for creating the recommended item detailed information can be acquired, display information may be created in association with the set identifier. .
The display total information storage unit 136 stores a plurality of display total information. FIG. 9 is a diagram illustrating a storage state of the display total information storage unit 136. Display aggregate information is a user identifier (user_i
d), item identifier (item_id), and browsing rate total value (view_total)
And the last display time information (last_disp_date) are associated with each other and stored in a table format as shown in FIG. The browsing rate total value is a value obtained by totaling the browsing rates to be described later, and the larger the number of items displayed as item detailed information on the terminal display unit 34 of the terminal device 3, the larger the value. Therefore, the larger the browsing rate total value, the higher the possibility that the user corresponding to the user identifier is browsing the item many times. The last display time information is time information when the item corresponding to the item identifier included in the display total information is last displayed on the terminal display unit 34 of the terminal device 3. (User_id,
The combination of item_id) is unique and does not allow duplication. The final display time information is the transmission time information in the present embodiment.

The information processing server control unit 11 performs overall control on each unit configuring the information processing server device 1. The information processing server control unit 11 includes a usage history registration unit 111, a display information registration unit 112, a recommended item selection unit 113, a display information aggregation unit 114, and a display order determination unit 11.
And 5.
The usage history registration unit 111 receives the usage history from the terminal device 3 via the network 2, and stores the received usage history in the usage history storage unit 133.
The display information registration unit 112 performs display information registration processing. In the display information registration process, display information is created in accordance with the storage format of display information based on the display history received from the terminal device 3 via the network 2, and the created display information is stored in the display information storage unit 135. Is stored.

The recommended item selection unit 113 performs a recommended item selection process for a recommendation target user who is a user to be recommended.
In the recommended item selection process, first, a recommendation value between the recommendation target user and each item is calculated.
Next, a plurality of recommended items are selected based on the calculated recommended value. Next, for each selected recommended item, recommended item information that associates at least the user identifier of the user to be recommended, the item identifier of the selected recommended item, and the recommended value is created. The recommended item information storage unit 134 stores all of the created recommended item information. Note that a recommended item selection method is a method of calculating a recommended value between a recommendation target user and each item and selecting a plurality of recommended items based on the calculated recommended value. Note that a set of items for which recommended values are calculated is a recommended candidate set.

In the recommended item selection process, a plurality of recommended item selection methods are prepared and a recommended item is selected for each recommended item selection method for a recommended target user, thereby recommending a recommended item for one recommended target user. Multiple sets of information can be created. At this time, the recommended item information storage unit 134 needs to use the second storage format in order to distinguish the set of selected recommended item information for each recommended item selection method. And after associating individual set identifiers for each recommended item selection method in advance, the user identifier of the user to be recommended,
The recommended item information that associates the set identifier associated with the recommended item selection method, the item identifier of the recommended item, and the recommended value may be stored in the recommended item information storage unit 134.

The recommended item selection process may be performed every time a recommended item acquisition request is received from the terminal device 3. In addition, for each predetermined timing predetermined by the service provider,
A user identifier of a user who satisfies a predetermined condition may be extracted, and a recommended item selection process may be performed for each extracted user identifier. As the predetermined timing, a predetermined time interval (for example, 2
Every 4 hours) or every time the usage history is received a certain number of times. Also,
The time interval may vary, such as every 3 hours from Monday to Friday, every 6 hours on Saturday, and every 12 hours on Sunday. Also, the time interval may be changed according to the season, such as shortening the time interval in summer and increasing the time interval in winter. In addition, the user who satisfies the predetermined condition may be a user who uses some item in a predetermined period such as one month retroactively from the present time, or may be a user who has viewed the recommended item within the predetermined period. The user may use a predetermined number (for example, five) or more items. Of course, all users who use the service may be used.

Here, an example of a recommended item selection method will be described.
First, using a usage history stored in the usage history storage unit 133, a similar user as a recommendation target user is selected. In order to select similar users, for example, there is a method of calculating similarity using a Jaccard coefficient and selecting similar users using the calculated similarity. Specifically, the item set I (ub) used by the recommendation target user ub and the item set I (us) used by another user us are items that exist in both I (ub) and I (us). The number of items existing in at least one of I (ub) and I (us) is | I (ub) ∪I (us) | Similarity sim using 1)
(Ub, us) is calculated. Then, a predetermined number of users or users whose calculated similarity is higher than a predetermined value are selected as similar users in descending order of the calculated similarity. Furthermore, if the degree of preference indicating the degree of preference for the user's item is calculated, the cosine distance, the Pearson product-moment correlation coefficient, and the like can be used for similarity calculation.

Next, the items used by the selected similar users are extracted as recommended candidate items. At this time, items used by the recommendation target user may be excluded. In the recommended item selection method, the set of recommended items extracted here is a recommended candidate set.
Next, a recommendation value with the recommendation target user is calculated for each extracted recommendation candidate item. In order to calculate the recommendation value, for example, the degree of similarity between the recommendation target user and the similar user can be used. Specifically, the recommendation candidate item is ir, and among the similar users of the recommendation target user ub, a set of users using the recommendation candidate item ir is Us (ub, ir), the recommendation target user ub and the recommendation candidate item ir are The similarity with the used similar user us ′ (∈Us (ub, ir)) is assumed to be sim (ub, us ′), and the recommended value v (ub, ir) is calculated using equation (2).

Then, the recommended item may be selected from the recommended candidate items based on the recommended value.
For selection of recommended items, recommended candidate items whose recommended values are larger than a predetermined value predetermined by the service provider may be selected. Alternatively, the service provider may select a predetermined number in descending order of the recommended value. Of course, all recommended candidate items may be selected as recommended items, but excluding recommended candidate items that have little relevance to the recommended user, for example, the lowest recommended value can be taken. It is desirable to do. In this way, by selecting recommended items based on the recommended values, only items that are highly recommended to the user can be set as recommended items.
The above is an example of the recommended item selection method. Of course, a recommendation item may be selected by calculating a recommendation value between the recommendation target user and the item using a method other than the above method.
When the recommended item information newly created is stored in the recommended item information storage unit 134, if the recommended item information created before the user to be recommended already exists, the recommended item created before The newly created recommended item information may be stored after deleting all the information. In addition, a section for storing previously created recommended item information may be prepared in the information processing server storage unit 13 to hold previously created recommended item information.
Note that the recommended item selection unit 113 in this embodiment has a role of a recommended value calculation unit and a similar user selection unit.
The display information totaling unit 114 performs display information totaling processing.
The display information aggregation process may be performed every time display information is stored in the display information storage unit 135. Further, similar to the recommended item selection process, it may be performed at a predetermined timing predetermined on the service providing side.

Two types of display information aggregation processing procedures will be described below.
First, the first display information aggregation process will be described with reference to FIG. The first display information tabulation process updates the display tabulation information using only newly registered display information, in other words, display information that has not been used in the previous processing.
First, the display information totaling unit 114 acquires all unprocessed display information from the display information storage unit 135 (step S101). In order to determine unprocessed display information, a process determination flag indicating whether the display information has been processed or not processed is added to the display information, and the determination may be made based on the flag. In this case, when registering in the display information storage unit 135, the display information processing determination flag is left unprocessed, and when it is acquired in this step, the display information processing determination stored in the display information storage unit 135 is determined. The flag may be updated from unprocessed to already processed. Further, the latest acquisition time information indicating the time when the previous process was performed may be stored in the information processing server storage unit 13 and only display information registered after the latest acquisition time information may be acquired as unprocessed display information. Good. In this case, after acquiring all unprocessed display information, the latest acquisition time information may be updated. When page information is included in the display information, it can be acquired only for unprocessed display information having page information that satisfies a predetermined condition. Specifically, for example, the processing may be performed by limiting the page information to display information that is recommended page information. Limited to the display information corresponding to the recommended page information,
By performing the processing described later, only the browsing status of recommended items can be aggregated.

Next, the display information totaling unit 114 selects one of the display information acquired in step S101, for example, in the order of acquisition (step S102).
Next, the display information totaling unit 114 calculates a viewing rate for each item identifier included in the display information selected in step S102 (step S103). Display information storage unit 1
When the storage format of 35 is the first storage format, one or more item identifiers are included in the display detailed information of the display information, so it is necessary to calculate a viewing rate for each item identifier. The browsing rate is a value that increases as the possibility that the user has browsed the item increases, and the property of the browsing rate also changes depending on the calculation method of the browsing rate.
Below, the calculation method of a browsing rate is demonstrated.
The first method for calculating the browsing rate is a method for uniformly assigning the same value (for example, “1”) to any item identifier. In this method, all displayed items are viewed by the user. At this time, the browsing rate total value, which is a value obtained by totaling the browsing rates, is proportional to the number of times the item is displayed on the terminal display unit 34 of the terminal device 3.

The second method for calculating the viewing rate is a method for calculating the browsing rate so that the higher the final display rank, the higher the value. Specifically, for example, when the final display order for the item i of the user u is rank (u, i), the browsing rate vr (u, i) is a constant α1 (> 0) and a constant α2.
Using (> 0) and the constant β1 (> = 0), the calculation can be performed using Equation (3).

In addition, the ceiling function ceil and 2 are set so that the viewing rate changes for every fixed final display order.
Using the integer N, the constant α3 (> 0), the constant α4 (> 0), and the constant β2 (> = 0), it can also be calculated by the equation (4). Moreover, as long as it decreases monotonously as in equation (5), the browsing rate calculation formula may be changed according to the final display order.

As described above, any function that monotonously decreases with respect to the final display order may be used. Further, as shown in FIG. 11, a correspondence table in which a numerical value that monotonously decreases in accordance with the final display order is associated with the final display order as a viewing rate is stored in the information processing server storage unit 13 in advance. You may make it acquire a browsing rate from a table | surface. The second method of calculating the browsing rate is a method that is more suitable when an item having a higher final display order is displayed at the top of the page displaying the item detail information. This is a method of calculating the viewing rate based on the assumption that the user is viewing the item displayed at the top of the page firmly, but the item at the bottom of the page is hardly viewed It is. At this time, the browsing rate total value tends to be larger as the number of times the item is displayed on the upper part of the page related to the item detailed information being displayed on the terminal display unit 34 of the terminal device 3 increases.

In addition, when page information is included in display information, the calculation method of the browsing rate to be used may be changed for each page information. For example, as shown in FIG. 12 (FIGS. 12A to 12C), three recommended pages are prepared for one service, corresponding to FIG. 12A in which the recommended page information is p1. In the recommended page, only 3 recommended items are displayed, and the recommended page corresponding to FIG. 12B in which the recommended page information is p2 is displayed in 20 recommended items arranged vertically. In the recommendation page corresponding to FIG. 12C in which p3 is p3, it is assumed that three rows of recommended items arranged side by side are displayed for 10 lines. At this time, since only three recommended items are displayed for the recommended page information p1, the browsing rate is calculated as “1” using the first method of calculating the browsing rate, and the recommended page information p2 Since 20 recommended items are displayed side by side, the calculation is performed using the formula (3) of the second method for calculating the browsing rate. For the recommended page information p3, 30 recommended items are displayed. Since there are 10 rows arranged side by side, N = 3 may be calculated using equation (4). In this way, by changing the formula for calculating the browsing rate according to the page information, depending on how the item is displayed,
The browsing rate can be calculated.

Further, as shown in FIG. 13, a correspondence table in which page information and constants are associated with each other is stored in advance in a page information correspondence table storage unit (not shown) of the information processing server storage unit 13. A constant may be acquired as a browsing constant according to page information, and the acquired browsing constant may be used as a browsing rate. At this time, the browsing rate of each item identifier contained in the display information which has the same page information becomes the same value. Further, the browsing rate may be calculated so as to monotonously decrease in accordance with the display order after setting the browsing constant as the maximum value of the browsing rate.
In addition, regarding the display information corresponding to the recommended page information, when the display information includes a set identifier, the method of calculating the browsing rate to be used may be changed for each set identifier.
The above is the description of the browsing rate calculation method.
Next, the display information totaling unit 114 performs display totalization registration update processing for updating and registering the display total information (step S104). The display aggregation registration update process will be described later.
Next, the unprocessed display information acquired by the display information totalization unit 114 in step S101 is
It is determined whether or not all selections have been made in step S102 (step S105). If all are selected, the process proceeds to step S106. If unselected items still remain, step S
Proceed to 102.
In step S106, the display information aggregation unit 114 deletes the display aggregation information to be deleted from the display aggregation information storage unit 136, and ends a series of processing from step S101 to step S106. What is to be deleted in step 106 is display aggregate information relating to the user's items that have not been displayed for a predetermined period or longer determined by the service provider. For this, it is only necessary to delete the period between the time at the time of processing and the final display time information included in the display total information over a predetermined period. If the display aggregation information is not deleted, step S106 may be omitted. Moreover, you may perform the process regarding step S106 for every predetermined | prescribed timing predetermined by the service provision side instead of performing by a 1st display information total process.

Here, the procedure of the display aggregation registration update process in step S104 will be described with reference to FIG.
First, the display information totaling unit 114 selects one item identifier of the browsing rate calculation target item whose browsing rate is calculated in step S103, for example, in the order of dictionary of item identifiers (step S201).
Next, the display information totaling unit 114 has display total information in which the user identifier included in the display information selected in step S102 matches the item identifier selected in step S201 in the display total information storage unit 136. It is determined whether or not (step S202). When it exists, it progresses to step S203, and when it does not exist, it progresses to step S204.
In step S203, the display information totaling unit 114 uses the display total information storage unit 136 to display the display total information that matches the user identifier included in the display information selected in step S102 and the item identifier selected in step S201. The browsing rate total value and the final display time information of the specified display total information are specified and updated as necessary. Next, the process proceeds to step S205.

There are the following two methods for updating the browsing rate aggregate value.
The first update method of the browsing rate total value is a method of adding the browsing rate calculated in step S103 to the browsing rate total value. At this time, the browsing rate total value is the sum of the browsing rates.

The second method of updating the browsing rate aggregate value is to calculate the update value of the browsing rate aggregate value using the average value of the browsing rate obtained by dividing the sum of the browsing rates by the number of display times and the display count, and the calculated update value Replace the browsing rate aggregate value with. Specifically, for example, the average value of the browsing rate for the item i of the user u is av
r (u, i), the number of times of display is c (u, i), and a constant δ (> 0) for adjusting the influence of the number of times of display is set, and then the browsing rate total value rt (u, i) ) May be calculated using equation (6).

When “0 <δ <1”, the influence of the number of times of display becomes smaller than that of the first update method of the browsing rate update, and when “δ> 1”, the display rate is displayed by the first update method of the browsing rate update. The influence of times increases. In the case of “δ = 1”, since it is equal to the sum of the browsing rates, the influence of the number of display times is equivalent to the first updating method of the browsing rate update. Since the second update method of the browsing rate total value requires the sum of the browsing rates and the number of times of display, the total of the browsing rates and the number of times of display can be associated with the display total information and stored in the display total information storage unit 136. It is necessary to. Each time the display summary information is updated, the browsing rate is added to the sum of the browsing rates, and “1” is added to the display count.

The final display time information is updated only when the display time information included in the display information selected in step S102 is newer than the final display time information of the specified display aggregation information. When updating, the last display time information is updated with the display time information included in the display information selected in step S102.
In step S204, the display information totaling unit 114 uses the user identifier included in the display information selected in step S102, the item identifier selected in step S201, and the browsing rate calculated in step S103 as the browsing rate total value. Display aggregation information that associates the display timing information included in the display information selected in step S102 with the final display timing information is registered in the display aggregation information storage unit 136. When the sum of the browsing rates and the display count are associated with the display total information, the sum of the browsing rates may be the browsing rate calculated in step S103, and the display count may be set to “1”.

In step S205, the display information totaling unit 114 determines whether or not all item identifiers included in the display information selected in step S102 have been selected in step S201.
If all are selected, the series of processing from step S201 to step S205 is terminated, and if there are still unselected items, the process proceeds to step S201.
The above is the description of the procedure of the display aggregation registration update process in step S104.
Since the first display information totalization process handles only unprocessed display information, the processing amount is small. For this reason, it is suitable for the case where it is performed every time display information is stored in the display information storage unit 135.
The above is the description of the procedure of the first display information aggregation process.

Next, the procedure of the second display information tabulation process will be described with reference to FIG.
First, the display information totaling unit 114 deletes all the display total information stored in the display total information storage unit 136 (step S301).
Next, the display information totaling unit 114 acquires all display information having display time information corresponding to a predetermined processing period from the display information storage unit 135 (step S302). The predetermined processing period may be a period that goes back a predetermined time (for example, one day, one week, one month, etc.) predetermined by the service provider from the time when the second display information aggregation processing is performed. Also in the second display information aggregation process, if the display information includes page information, only the display information satisfying the predetermined condition is acquired and the subsequent processes are performed as in the case of the first display aggregation process. You may go.

In step 102, if display information is selected from the display information acquired in step S302, the subsequent procedure is step S102 of the first display information aggregation process.
To step S105.
In the second display information totaling process, since the display total information is created from scratch, the processing amount is large, but the display total information can be generated using only display information corresponding to a predetermined processing period. Note that the second display information tabulation process has a large amount of processing, and is preferably performed at predetermined timings.
The above is the description of the procedure of the second display information aggregation process. The display information totaling unit 114 serves as a browsing rate total value calculation unit in the present embodiment.

Upon receiving a recommended item acquisition request from the terminal device 3, the display order determination unit 115 performs display order determination processing. Moreover, the item detailed information transmission process which transmits item detailed information to the terminal device 3 is performed. When the recommended item selection processing is performed every time the recommended item selection unit 113 receives a recommended item acquisition request from the terminal device 3, the display order determination unit immediately after the recommended item selection processing by the recommended item selection unit 113. 115 may perform display order determination processing.

First, the procedure of the display order determination process will be described with reference to the flowchart of FIG.
First, the display order determination unit 115 receives a recommended item acquisition request from the terminal device 3 via the network 2 (step S401).

Next, the display order determination unit 115 acquires recommended item information from the recommended item information storage unit 134 (step S402). When the recommended item information storage unit 134 is in the first storage format, all the recommended item information including the user identifier included in the recommended item acquisition request received in step S401 may be acquired. When the recommended item information storage unit 134 is in the second storage format, the recommended item information including the user identifier included in the recommended item acquisition request received in step S401 and the collective identifier acquired using the recommended item acquisition request is displayed. All you need to get is In order to obtain the collective identifier, if the recommended item acquisition request received in step S401 includes the collective identifier, it may be obtained as it is, or the recommended item acquisition request received in step S401 may be included in the recommendation page. If information is included and set determination information in which recommended page information and a set identifier are associated with each other is stored in the information processing server storage unit 13, the information may be acquired from the information processing server storage unit 13. Further, the information processing server storage unit 13 holds information regarding the page currently being browsed by the user corresponding to the user identifier included in the recommended item acquisition request received in step S401, and the information processing server storage unit Even when the set determination information is stored in 13, the information processing server storage unit 13
You can get more.

Next, when the display order determining unit 115 has display total information that matches one of the combinations of the user identifier and the item identifier included in the recommended item information acquired in step S402 from the display total information storage unit 136. Are all acquired (step S403).
Next, the display order determination unit 115 calculates a priority for each item identifier included in the recommended item information acquired in step S402 (step S404).

Here, a method for calculating the priority will be described. Of the recommended item information, step S4.
In 03, the browsing rate total value of the recommended item information for which the corresponding display total information could not be acquired is regarded as “0”.

The priority has such a property that the larger the recommended value, the larger the value, and the smaller the browsing rate total value, the larger the value. Specifically, for example, the recommended item ir of the user u
When the recommended value of v is (v, u, ir) and the browsing rate total value rt (u, ir), the constant γ1 (
> 0), constant γ2 (> 0), constant γ3 (> 0), constant γ4 (> 0), and constant α5 (> 0), and the priority pr (u, ir) is given by equation (7) And the equation (8).

In addition, the recommended item information acquired in step S402 is rearranged in the descending order of the recommended value, and the ranking given in order from the first is used as the recommended ranking, and the recommended ranking is used instead of the recommended value. It is also possible to calculate the priority so that the higher the value is, the larger the value becomes. Specifically, for example, the recommendation rank rankr (u, i) of the recommended item ir of the user u
r), a constant γ5 (> 0), a constant γ6 (> 0), a constant γ7 (> 0), and a constant γ8 (
> 0) and the constant α6 (> 0), the priority pr (u, ir) is calculated using the equations (9) and (10).
).

Of course, the larger the recommended value is, the larger the value is. The priority may be calculated using any calculation formula as long as the calculation formula has any one of the properties of increasing the value as the value decreases.

Next, the display order determination unit 115 assigns the display order from the first to the item identifier included in the recommended item information acquired in step S402 in order from the highest priority calculated in step S404 ( Step S405). If items that have been used in the past are excluded from the recommended items, the display order of items that have been recommended many times but have not been used will be lowered by assigning the display order based on priority. Can do.

Next, the display order determining unit 115 receives the content information storage unit 132 from step S402.
Item information corresponding to the item identifier included in the recommended item information acquired in step S405 is acquired, sorted in the order of display order assigned in step S405, and then sorted to the terminal device 3 through the network 2 in the order of display order. Then, a set of item information associated with the display order is transmitted as recommended item detailed information (step S406), and a series of processing from step S401 to step S406 is completed. At this time, if the acquisition limit number is set in the recommended item acquisition request received in step S401, a set of item information up to the acquisition limit number in order from the top of the display order is set as the recommended item detailed information.

When the terminal device 3 divides and acquires the recommended item detailed information, the recommended item detailed information created by performing the display order determination process is stored in the information processing server storage unit 13 and then the terminal device. 3 of the stored recommended item detailed information in response to the request 3.
What is necessary is just to transmit the item information of the display order of the predetermined range designated by. For example, first, when receiving a request for item information related to recommended items from the display order 1st to 5th from the terminal device 3, the recommended item detailed information created by performing the display order determination process,
After storing the information in the information processing server storage unit 13, the item information regarding the recommended items from the first display position to the fifth display order, which is a part of the created recommended item detailed information, is transmitted to the terminal device 3. Next, the display order 6 from the terminal device 3 corresponding to the recommended item detailed information created earlier.
When the request for the item information regarding the recommended items from the first to the 20th is received, the display order determination process is not performed, and the recommended items from the display rank 6th to the 20th, which are a part of the recommended item detailed information created earlier. The item information may be transmitted to the terminal device 3.

As described above, using a priority having a property that the higher the recommended value indicating the degree of recommendation for the user's item, the higher the value, and the lower the browsing rate total value indicating the degree that the user has browsed the item. By determining the display order, items that are not often viewed tend to be highly recommended items, so the recommendation results can be presented in a variety of ways while maintaining the accuracy of the recommendation results. Therefore, it is possible to present a recommended item that ensures both the reliability of the recommended item and the maintenance of interest. The same effect can be obtained when the priority is calculated using the recommendation ranking and the browsing rate total value.

In addition, since the browsing rate total value increases in proportion to the number of times the recommended item is displayed, the recommended item is frequently replaced for a user with a high number of recommended item display times, but the number of recommended item display times is low. In contrast, the ratio of replacing the recommended items is automatically adjusted according to the browsing status of the recommended items of the user so that the recommended items are not frequently replaced.
In addition, by adjusting the constant of the browsing rate calculation formula,
It is possible to adjust how much recommended items are replaced.
The above is the description of the procedure of the display order determination process.

Next, item detailed information transmission processing will be described.
The item detailed information transmission process is a process of transmitting item detailed information other than the recommended item detailed information to the terminal device 3. First, the item information necessary for the page displayed on the terminal display unit 34 of the terminal device 3 All are acquired from the information storage unit 132. Next, item detailed information is created using the acquired item information. And the created item detail information,
It transmits to the terminal device 3.
The above is the description of the item detailed information transmission process.

An example of a series of procedures in which the terminal device 3 acquires the recommended item detailed information and transmits the display history will be described with reference to the flowchart of FIG.
First, the terminal control unit 31 of the terminal device 3 performs a recommended item acquisition request transmission process, and transmits a recommended item acquisition request to the information processing server device 1 through the network 2 (step S501).
Next, when the recommended item selection unit 113 of the information processing server device 1 receives a recommended item acquisition request from the terminal device 3 through the network 2, the recommended item is acquired with respect to the user identifier included in the received recommended item acquisition request. A selection process is performed (step S502).

Next, the display order determination unit 115 of the information processing server apparatus 1 performs a display order determination process on the user identifier included in the recommended item acquisition request received in step S <b> 502. Then, the recommended item detailed information corresponding to the recommended item acquisition request received in step S502 is transmitted (step S503).
Next, when the terminal control unit 31 of the terminal device 3 receives the recommended item detailed information corresponding to the recommended item acquisition request transmitted in step S501 from the information processing server device 1 through the network 2, the recommended item detailed information is displayed. Processing is performed (step S504).

Next, the item information displayed on the terminal display unit 34 of the terminal device 3 among the item information of the recommended items included in the recommended item detailed information received by the terminal control unit 31 of the terminal device 3 in step S504. Then, display history transmission processing is performed, and the display history is transmitted to the information processing server device 1 through the network 2 (step S505).
Next, when the display information registration unit 112 of the information processing server device 1 receives a display history from the terminal device 3 through the network 2, a display information registration process is performed on the received display history (step S506). At this time, the display information created based on the display history received in step S506 is stored in the display information storage unit 135 as unprocessed display information.

Next, the display information totalization unit 114 of the information processing server device 1 performs processing on unprocessed display information.
Display information totaling processing is performed (step S507), and a series of processing from step 501 to step S507 is terminated.
If the recommended item selection process is performed at every predetermined timing, step S50 is performed.
The process 2 may be skipped. In addition, when the display information aggregation process is performed at every predetermined timing, the process of step S507 may be skipped.
The above is an example of a series of procedures for acquiring recommended item detailed information and transmitting a display history.

In the present embodiment, the display history is received from the terminal device 3, but the item displayed on the terminal display unit 34 of the terminal device 3 among the item detailed information without receiving the display order, If the final display order of the items to be displayed is clear, instead of receiving the display history from the terminal device 3, the display history is created by the display order determination process in the display order determination unit 115 or the item detailed information. Also good. At this time, the display information registration unit 112 may perform display information registration processing on the display history created by the display order determination unit 115.

Further, in the present embodiment, the display information storage unit 135 is prepared in the information processing server device 1, but instead of being prepared in the information processing server device 1, it is prepared in a terminal storage unit (not shown) of the terminal device 3 and described above. When the display information is used in this process, it may be appropriately acquired from the terminal storage unit of the terminal device 3. At this time, display information registration processing performed by the display information registration unit 112 needs to be performed by the terminal control unit of the terminal device 3.

Further, when a display information storage unit exists in a terminal storage unit (not shown) of the terminal device 3, a display total information storage unit is further prepared in the terminal storage unit, and then the display control unit 31 displays the display information total unit. The display information aggregation process in 114 may be performed, and the display aggregation information may be transmitted to the information processing server device 1 after the process is completed. In this case, the display information totaling unit 114 needs to store the display total information received from the terminal device 3 in the display total information storage unit 136. In addition, a display information storage unit and a display total information storage unit are prepared in a terminal storage unit (not shown) of the terminal device 3, and item information and recommended values of recommended items are acquired from the information processing server device 1. The terminal control unit 31 of the terminal device 3 may calculate the priority.

Moreover, even if it is not the recommended item selected for every user, if the presentation degree which shows the degree presented for every item is set, it can consider that a presentation degree is the recommended item which makes a recommendation value, and can perform the said process. it can.

The form of the service in which the effect by the first embodiment of the present invention is particularly high will be described below.
When a recommended item selection process with a large amount of processing such as a collaborative filtering method is applied in a service frequently accessed by many users, the recommended item selection process for the user is performed each time the user accesses Since the amount of processing increases and the processing cannot be completed, the recommended item selection processing is often performed for users who satisfy a predetermined condition at every predetermined timing. In such a service, by performing the display order determination process in the present embodiment, the display order of the recommendation results can be switched according to the user's browsing situation without frequently performing the recommended item selection process.

<Second Embodiment>
Below, the 2nd Embodiment of this invention is described in detail using figures. Second of the present invention
This embodiment is a method of using a browsing rate total value when calculating a recommended value for a recommended item.

The configuration of the entire system in the second embodiment of the present invention is the same as that of the first embodiment of the present invention except that the information processing server device 4 is used instead of the information processing server device 1.

The network 2 and the terminal device 3 are the same as those in the first embodiment of the present invention.
The information processing server device 4 is a device that exchanges various information with the terminal device 3 via the network 2. The information processing server device 4 can be implemented as software (program) processing using a computer including a CPU, RAM, ROM, hard disk drive (HDD), network interface, and the like. In addition, you may comprise the information processing server apparatus 4 using a some computer. For example, to load balance
The distributed processing may be performed using a plurality of computers that perform processing corresponding to each unit of the information processing server device 4. Alternatively, the distributed processing may be performed in such a manner that the processing of a part of the information processing server device 4 is performed by a computer and the processing of other units is performed by another computer.

FIG. 18 is a configuration diagram of the information processing server device 4 in the present embodiment. The information processing server device 4 in this embodiment includes an information processing server control unit 41 and an information processing server communication unit 12.
And the information processing server storage unit 13.
The information processing server communication unit 12 and the information processing server storage unit 13 are the same as those in the first embodiment of the present invention.

The information processing server control unit 41 performs overall control on each unit configuring the information processing server device 4. The information processing server control unit 41 includes a usage history registration unit 111, a display information registration unit 112, a recommended item selection unit 413, a display information aggregation unit 114, and a display order determination unit 41.
And 5.
The usage history registration unit 111, the display information registration unit 112, and the display information totalization unit 114 are the same as those in the first embodiment of the present invention.

Similar to the first embodiment of the present invention, the recommended item selection unit 413 performs a recommended item selection process for the recommendation target user, but instead of calculating a recommended value when selecting a recommended item, A second recommended value is calculated, and the calculated second recommended value is used for selecting recommended items. In order to calculate the second recommended value, the recommended value in the first embodiment of the present invention is used as a reference recommended value. The larger the reference recommended value, the larger the value, and the smaller the browsing rate total value, the more the value. Is calculated to be larger. That is, the second recommended value has the same property as the priority in the first embodiment of the present invention.

Therefore, the reference recommended value for the recommendation candidate item ir ′ of the recommendation target user u is set to vb (u
, Ir ′), and the browsing rate total value rt (u, ir ′) for the recommendation candidate item ir ′ of the recommendation target user u is used as the second recommended value v ′ (u, ir ′). First of the invention
P (u, ir) in the formula (7) and the formula (8) which are the calculation formulas of the priority in the embodiment of v ′ (
u, ir ′), v (u, ir) to vb (u, ir ′), and rt (u, ir) to rt (u
, Ir ′) can be calculated.

In addition, the recommendation candidate items are sorted in descending order of the standard recommendation value, and the recommendation candidate ranks are assigned in order from the first place. , Ir ′), and p (u, ir) in the formulas (9) and (10), which are the priority calculation formulas in the first embodiment of the present invention, is changed to v ′ (u, ir ′), and rankr (U,
The second recommended value can also be calculated by replacing ir) with rankr ′ (u, ir ′) and rt (u, ir) with rt (u, ir ′). The browsing rate total value of the recommendation candidate item of the recommendation target user is stored in the display total information storage unit 136 with the user identifier of the recommendation target user,
If there is display aggregate information that matches the combination of item identifiers of recommended candidate items,
What is necessary is just to acquire from matching display total information. If there is no matching display total information, the browsing rate total value may be set to “0”.
As described above, by using the browsing rate aggregate value for the calculation of the second recommended value, it is easy to exclude items displayed as recommended items repeatedly from the set of recommended items for the recommendation target user.

Also, recommended users who browse recommended items frequently but rarely browse or use items other than items used in the past, or whose interest in items hardly changes. When the recommended item is calculated using the value,
The recommended value for the recommended candidate item hardly changes, and almost the same item may be selected every time as the recommended item. The recommended item set hardly changes. Even if you select another item from the recommended item set each time, if you browse the recommended item many times, the items recommended in the past will be displayed many times as a result. May become unable to maintain interest in the user's recommendation. For such a recommendation target user, selecting the recommended item using the second recommended value using the browsing rate total value increases the change in the second recommended value for the recommended candidate item. Each time the item set selects a recommended item, the item is replaced, and the user's interest in the recommendation can be maintained.

As in the first embodiment of the present invention, the display order determination unit 415 performs a display order determination process when receiving a recommendation item acquisition request from the terminal device 3, but the display order determination process is different.

The display order determination process in this embodiment omits the processes from step S403 to step S405 of the display order determination process of the first embodiment of the present invention, and in step S406,
The priority may be calculated so that the higher the second recommended value is, the higher the value is, and the display order may be given in order of the calculated priority. Since the priority in this embodiment also has the same properties as the priority in the first embodiment of the present invention, the display order given here is the same as in the first embodiment of the present invention. Of the items that are not displayed as recommended items,
The higher the recommendation level, the higher the ranking.

An example of a series of procedures in which the terminal device 3 acquires recommended item detailed information and transmits a display history in the present embodiment is step S50 of the procedure in FIG. 17 in the first embodiment of the present invention.
The process 2 may be replaced with the recommended item selection process in the present embodiment, and the process in step S503 may be replaced with the display order determination process in the present embodiment.

In the present embodiment, since the recommended value becomes the reference recommended value and the second recommended value becomes the priority, the recommended item selection unit 413 and the display order determination unit 415 cooperate to display the display order determination unit in the present embodiment. Will work.

The form of the service in which the effect by the second embodiment of the present invention is particularly enhanced will be described below.
In this embodiment, every time a user accesses, a recommended item selection process for the user can be performed, so that any user can extract more recommended candidate items than the number of recommended items. If the recommended item selection method is used, the items constituting the set of recommended items change according to the number of times of display, so that the user's interest in the recommended items can be maintained.

In addition, a user who satisfies a predetermined condition at every predetermined timing even in a service in which the recommended item selection process for the user is performed every time the user accesses, and the processing amount becomes too large to be processed. By performing the recommended item selection process, the recommended item set of the recommendation target user with a large number of display times changes greatly at every predetermined timing, so that the user's interest in the recommended item can be maintained. Further, in the recommended item detailed information transmission process, when the display order is given so that the sorting order of the recommended items is changed every predetermined number of times, the recommended item set of the recommended target users having a large number of display times is set at every predetermined timing. It is possible to provide a recommended item that changes greatly and the order of the recommended items changes every time it is displayed a predetermined number of times.

<Third Embodiment>
Below, the 3rd Embodiment of this invention is described in detail using figures. Third of the present invention
The embodiment is a method of creating display aggregation information using display information to be described later when a display history is not transmitted from a terminal device.
The configuration of the entire system in the third embodiment of the present invention is the same as that of the first embodiment of the present invention except that the information processing server device 5 is used instead of the information processing server device 1.
The network 2 is the same as that of the first embodiment of the present invention.
The terminal device 3 has the same configuration as that of the terminal device 3 in the first embodiment of the present invention, but the processing performed by the terminal control unit 31 is different.
The terminal control unit 31 in the present embodiment performs the same processing as the terminal control unit 31 in the first embodiment of the present invention regarding the usage history transmission process, the recommended item acquisition request transmission process, and the item detailed information display process. Yes, but no display history transmission processing.

The information processing server device 5 is a device that exchanges various information with the terminal device 3 via the network 2. The information processing server device 5 can also be implemented as software (program) processing using a computer including a CPU, RAM, ROM, hard disk drive (HDD), network interface, and the like. In addition, you may comprise the information processing server apparatus 5 using a some computer. For example, to load balance
The distributed processing may be performed using a plurality of computers that perform processing corresponding to each unit of the information processing server device 5. Further, the distributed processing may be performed in such a manner that the processing of a part of the information processing server device 5 is performed by a computer and the processing of the other part is performed by another computer.

FIG. 19 is a configuration diagram of the information processing server device 5 in the present embodiment. The information processing server device 5 in this embodiment includes an information processing server control unit 51 and an information processing server communication unit 12.
And an information processing server storage unit 53.
The information processing server communication unit 12 is the same as that of the first embodiment of the present invention.
The information processing server storage unit 53 stores various data using a storage device such as a memory or an HDD. The information processing server storage unit 53 includes a user information storage unit 131, an item information storage unit 132, a usage history storage unit 133, a recommended item information storage unit 134, a display information storage unit 535, and a display total information storage unit 136. It consists of. A user information storage unit 131;
Item information storage unit 132, usage history storage unit 133, and recommended item information storage unit 134
The display total information storage unit 136 is the same as that of the first embodiment of the present invention.

The display information storage unit 535 stores a plurality of display information created based on the item detailed information transmitted to the terminal device 3. A method for creating display information in the present embodiment will be described later. FIG.
0 is a diagram showing a storage state of display information. The display information is a user identifier (user_id
) And transmission time information (sent_date) which is time information when the item detailed information is transmitted
And display method information (disp_type) are associated with each other and stored in a table format as shown in FIG. The display method information is information for specifying information related to items constituting the item detailed information and information such as how the item detailed information is displayed. The information on how the information is displayed is, for example, information indicating how many items of the item detailed information are displayed, page information regarding a page on which the item detailed information is displayed, and the like. In the present embodiment, the display method information is the browsing specific information in the present embodiment, and the display information storage unit 535 is the browsing information storage unit in the present embodiment.

The information processing server control unit 51 performs overall control on each unit configuring the information processing server device 5. The information processing server control unit 51 includes a usage history registration unit 111, a recommended item selection unit 113, a display information aggregation unit 514, and a display order determination unit 515.
The usage history registration unit 111 and the recommended item selection unit 113 are the same as those in the first embodiment of the present invention.

When receiving a recommended item acquisition request from the terminal device 3, the display order determination unit 515 performs display order determination processing. Also, an item detail information transmission process is performed. Recommended item selection part 1
13 performs the recommended item selection process every time a recommended item acquisition request is received from the terminal device 3, the display order determination unit 515 determines the display order immediately after the recommended item selection process by the recommended item selection unit 113. Processing may be performed.

First, the procedure of the display order determination process in this embodiment will be described with reference to the flowchart of FIG.
A series of processing from step S401 to step S406 is the same as that of the first embodiment of the present invention. When the processing up to step S406 ends, the process proceeds to step S707.

In step S707, the display order determination unit 515 creates display format information. In order to create the display format information, sufficient information is required to specify how to display on the terminal display unit 34 of the terminal device 3. For example, a page for displaying multiple item detailed information If there is a display format defined for each page that displays item detail information, a page identifier that uniquely identifies the page that displays item detail information, such as the URL of the page or an identifier associated with the page May be created as display format information. The page identifier may be associated with the recommended item acquisition request received at step S401, or stored in the information processing server storage unit 53 in association with the user identifier included in the recommended item acquisition request received at step S401. In addition, the information may be acquired from the information processing server storage unit 53. In addition, for a plurality of display formats used in the service, a display format identifier that uniquely identifies the display format may be set, and the display format identifier may be used as display format information. Similar to the page identifier, the display format identifier may be associated with the recommended item acquisition request received in step S 401, or may be held in the information processing server storage unit 53.

Further, instead of the information specifying the display format, for example, the top 10 display rankings are acquired and displayed as they are, or the top 20 display rankings are acquired, and 1 is randomly selected from the acquired 20 pieces.
Information indicating the display format directly, such as displaying zero, may be used as the display format information. Also,
Information that associates the item identifiers of all the items that are actually displayed and the final display order may be used as the display format information. In addition, when the recommended item information storage unit 134 is in the second storage format, it is also necessary to associate a set identifier that specifies a set of recommended item information that is a generation source of the recommended item detailed information with the display format information. However, when the set identifier can be specified by other information such as a recommended page identifier, information for specifying the set identifier may be associated with the display format information instead of the set identifier.

Next, the display order determination unit 515 transmits to the display information storage unit 535 the user identifier included in the recommended item acquisition request received in step S401 and the time information when the recommended item detailed information is transmitted to the terminal device 3. Display information in which the time information is associated with the display format information created in step S707 is stored (step S708), and the series of processing ends.

In addition, although the process from step S707 to step S708 was demonstrated in the procedure performed after the process of step S406, as long as it is after step S401, it may be performed at any timing, and from step S402 to step S406. It may be performed in parallel with the processing.
The above is the description of the procedure of the display order determination process in the present embodiment.

Next, item detailed information transmission processing in the present embodiment will be described. The item detailed information transmission process in the present embodiment transmits the created item detailed information to the terminal device 3 in the same manner as the item detailed information transmission process in the first embodiment of the present invention. Display information is created, and the created display information is stored in the display information storage unit 535 in the same manner as the processing from step S707 to step S708 of the display order determination process.

The display information totaling unit 514 performs display information totaling processing.
The display information totaling process in the present embodiment may be performed every time display information is stored in the display information storage unit 535. Further, similar to the recommended item selection process, it may be performed at a predetermined timing predetermined on the service providing side.

Two types of display information aggregation processing procedures in this embodiment will be described below.
First, the first display information aggregation process in the present embodiment will be described with reference to FIG.
The first display information tabulation process in this embodiment is the first display information in the first embodiment of the present invention.
Since the process of step S103 of the display information totaling process of the above may be replaced with the process of step S803, the process of step S803 will be described.
In step S803, the display information totaling unit 514 uses the display information selected in step S802 to extract an item that is a target of browsing rate calculation (viewing rate calculation target item).
A browsing rate is calculated for each item identifier of the extracted browsing rate calculation target item.

Here, a method for extracting the item identifier of the browsing rate calculation target item will be described.
The method of extracting the item identifier of the browsing rate calculation target item differs depending on the type of display format information included in the display information selected in step S102.
When the item identifier of the displayed item is included in the display format information, the item identifier may be extracted as the item identifier of the browsing rate calculation target item.
Further, when the display format information is related to the recommended page information, even if the item identifier of the displayed recommended item is not included, the recommended item information stored in the recommended item information storage unit 134 and the display format information If the item identifier of the recommended item can be specified including the order displayed on the terminal display unit 34 of the terminal device 3, the specified item identifier may be extracted as the item identifier of the browsing rate calculation target item. At this time, past recommended item information is stored in the recommended item information storage unit 134 as necessary so that the recommended item information corresponding to the transmission time information included in the display information selected in step S102 can be acquired. Add an area to keep. Further, when the item identifier of the recommended item cannot be specified including the order displayed on the terminal display unit 34 of the terminal device 3, it corresponds to the user identifier and the transmission time information included in the display information selected in step S102. What is necessary is just to extract as an item identifier of a browsing rate calculation object item from the collection of recommended item information to perform.

In addition, when the recommended item information storage unit 134 is in the second storage format, when using the recommended item information stored in the recommended item information storage unit 134, information for specifying the set identifier is required. When creating the display format information that is a part of the display information in step S707 of the display order determination process, it is necessary to associate information for specifying the set identifier.

Next, a method for calculating the browsing rate will be described. The browsing rate calculation method can be calculated using the first method of browsing rate calculation according to the first embodiment of the present invention.

Moreover, when the last display order for every item can be specified, it can calculate using the 2nd method of browsing rate calculation.
In addition, when the terminal device 3 uses the specific rule and the random sort is performed, the item identifier and the final display order of the displayed item cannot be specified, but the specific rule can be determined. The following two calculation methods can be used.

The third method of calculating the browsing rate is a method of calculating a display probability estimated value that is an estimated value of a display probability that is a probability that an item is displayed as a browsing rate. For example, when the display probability estimated value is selected and displayed with 10 specific rules from 20 items in the terminal device 3, for example, the specific rule is randomly selected from up to 20 items with equal probability. If so, the display probability estimated value is “10/20 = 0.5”. Moreover, when a specific rule selects and displays 10 items with random bias from 20 items in the terminal device 3,
By performing 1000 trials using the same specific rule as the terminal device 3 and counting the number of display times in 1000 times for each item as shown in FIG. A probability estimate may be calculated. In either case, the display probability estimation value of items whose display rank is 21st or higher is “0”. The third method of calculating the browsing rate is
By estimating the display probability when the actually displayed item is not known, it is possible to increase the browsing rate of items that are likely to be actually displayed. Of course, when the display probability can be directly acquired, the display probability may be directly treated as the display probability estimated value.

The fourth method for calculating the browsing rate is to estimate the expected value of the final display order assigned to the item when the specific rule is biased random in the terminal device 3, and the estimated final display order This is a method of calculating the browsing rate so that the smaller the expected value, the larger the expected value. In addition,
The expected value of the final display order is, for example, for each item, the probability that the item will be first in the final display order is “0.2”, and the probability that the final display order is second is “0.1”. This can be calculated when the display order probability, which is the probability for each final display order, is known. Specifically, there are M items, and the probability that the display order k ranks in the item i of the user u is p.
When (u, i, M, k) is assumed, the expected value ev (u, i, M) of the final display order is expressed by the equation (11).
).

Even when the display order probability for each final display order is not known, for example, FIG.
As described above, using the same specific rule as that of the terminal device 3, the 20 items are sorted 1000 times, the average rank is calculated for each item, and the calculated average rank is the expected value of the final display rank. It can be a value. The average rank for an item is the display rank that is an estimate of the display rank probability for each final display rank by counting the number of times that rank was reached for each final display rank and dividing the counted number by the number of trials. It is also possible to calculate the expected value of the final display rank by calculating the probability estimate and substituting the display rank probability of Expression (11) with the display rank probability estimate. Thus, for each item, the expected value ev (u, i,
If M) is estimated, the viewing rate can be calculated by an equation in which rank (u, i) in equation (3) is replaced by ev (u, i, M). Of course, as long as the expected value of the final display order becomes smaller, the browsing rate may be calculated by other methods as long as it becomes a larger value. The fourth method for calculating the browsing rate is to use the expected value of the final display order when the actually displayed item is not found, so that the terminal display unit 34 of the terminal device 3 actually uses the expected value at the top of the page. The browsing rate of items that are likely to be displayed can be increased.

The above is the method of calculating the browsing rate that can be used when the item identifier and final display order of the displayed item cannot be specified, but the specific rule can be determined.
The above is the description of the procedure of the first display information aggregation process.

Next, the procedure of the second display information tabulation process will be described with reference to FIG.
The second display information tabulation process in the present embodiment is the second display information aggregation process in the first embodiment of the present invention.
The processing in step S103 of the display information aggregation processing may be replaced with the processing in step S803. Regarding step S803, the same process as step S803 in the first display information totaling process in the present embodiment may be performed.
The above is the description of the procedure of the second display information aggregation process.
It should be noted that the display probability estimation information for estimating the display probability estimation value and the display rank estimation information for estimating the display rank estimation value are included in the display format information. Further, the display order probability may be treated as a display order probability estimate.

The form of the service in which the effect by the third embodiment of the present invention is particularly high will be described below.
In a service where a large number of users frequently access and the load on the network is high, the display history is often not transmitted from the terminal device 3 in order to suppress the load on the network. Even in such a case, in this embodiment, the display order of recommendation results can be changed according to the user's browsing situation by using the display information.

Note that, in the third embodiment of the present invention, similarly to the second embodiment of the present invention, the browsing rate total value can be used when calculating the recommended value for the recommended item. in this case,
In the display order determination process of the display order determination unit 515, steps S403 to S405 are performed.
In step S406, the priority is calculated so that the higher the recommended value, the higher the value, and the display order may be given in the order of the calculated priority. In addition, after giving the display order in step S406, the item information sorted in the order of the display order included in the recommended item detailed information is re-sorted with randomness using a specific rule, It can also be displayed on the terminal display unit 34 of the terminal device 3 in the sorted order.
In this case, the third method or the fourth method of calculating the browsing rate is used when calculating the browsing rate, as in the case where the terminal device 3 uses the specific rule and the random sort is performed. Can do.

Using the present invention, by determining the display order of the recommended item detailed information corresponding to the recommended item acquisition request in descending order of priority using the recommended value or the recommended order and the browsing rate total value,
The accuracy of the entire recommendation result can be maintained, and a variety of display results can be presented. Therefore, it is possible to provide a recommended item that ensures both the reliability of the user's recommended item and the maintenance of interest. Can be promoted.

1, 4, 5 Information processing server device (information processing device)
2 Network 3 Terminal device 31 Terminal control unit 32 Terminal communication unit 33 Terminal input unit 34 Terminal display unit 11, 41, 51 Information processing server control unit 111 Usage history registration unit 112 Display information registration unit 113, 413 Recommended item selection unit 114, 514 Display information totaling unit (viewing rate total value calculation unit)
115, 415, 515 Display order determination unit 12 Information processing server communication unit 13, 53 Information processing server storage unit 131 User information storage unit 132 Item information storage unit 133 Usage history storage unit 134 Recommended item information storage unit 135, 535 Display information storage (Browsing information storage)
136 Display summary information storage

Claims (10)

Calculating an aggregate value with monotonically decreasing characteristic for ranking the number of times that one item information is provided or displayed regarding one use mainly becomes a higher value more, and relating to the provision or display of the one item information A total value calculation unit to
The higher the recommended value indicating the recommendation level of the one item information and the lower the total value, the higher the priority, the higher the priority, for each combination of the one usage subject and each item information that is a selection candidate. A priority calculation unit for calculating
An information selection unit that selects item information provided to the one user based on the priority;
An information processing apparatus comprising: The aggregate value calculation unit has a high provision order when the one item information is provided to the one user or a display order in which the one item is displayed on the terminal device related to the one user. Then, using the rank value that becomes a small value, the aggregate value is calculated so as to have a monotonous decrease characteristic with respect to the rank value.
The information processing apparatus according to claim 1. The total value calculation unit selects one function from functions having different monotonic decreasing characteristics based on the type of page on which the one item information is provided, and uses the one function to calculate the total value To calculate,
The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. The total value calculation unit treats the ranks related to items arranged in the same row as the same rank in the page provided with the one item information, and the item arranged in the upper row of the page has its value The aggregate value is calculated so that the
The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. The aggregate value calculation unit, a first value is calculated, by adding the only pre Kikai number first value monotonously decreases to rank related to the provision or display of the one item information, the aggregate value To calculate,
The information processing apparatus according to claim 1 , wherein the information processing apparatus is an information processing apparatus. The aggregate value calculating unit uses Oite the total value, for the rank related to the provision or display of the one item information, a parameter for adjusting the size of the previous Kikai number of influential, the aggregate value calculate,
The information processing apparatus according to any one of claims 1 to 5, characterized in that. The total value calculation unit sets the parameter to a value greater than 0 , calculates a first value that monotonously decreases with respect to the rank related to the one item information, for each provision of the one item information, by using the average value of, before Kikai number and base power value for the index of the parameter, a multiplication value between, calculates the total value,
The information processing apparatus according to claim 6 . The aggregate value calculation unit, with an increase in pre-Kikai number, as nonlinearly value increases, and calculates the total value,
The information processing apparatus according to any one of claims 1 to 7, characterized in that. An information processing method executed by one or more computers,
Calculating an aggregate value with monotonically decreasing characteristic for ranking the number of times that one item information is provided or displayed regarding one use mainly becomes a higher value more, and relating to the provision or display of the one item information A total value calculation step to be performed;
The higher the recommended value indicating the recommendation level of the one item information and the lower the total value, the higher the priority, the higher the priority, for each combination of the one usage subject and each item information that is a selection candidate. A priority calculating step to calculate
An information selection step of selecting item information provided to the one user based on the priority;
An information processing method comprising: One or more computers,
Calculating an aggregate value with monotonically decreasing characteristic for ranking the number of times that one item information is provided or displayed regarding one use mainly becomes a higher value more, and relating to the provision or display of the one item information A summary value calculation unit,
The higher the recommended value indicating the recommendation level of the one item information and the lower the total value, the higher the priority, the higher the priority, for each combination of the one usage subject and each item information that is a selection candidate. A priority calculation unit for calculating
An information selection unit that selects item information provided to the one user based on the priority;
An information processing program that functions as a computer program.

Images