Item list providing system, server, terminal, item list providing method, and program

Abstract

In an item list providing system (100) provided with a server (10) and a terminal (20), the server (10) and the terminal (20) execute, in cooperation with each other, processes of: generating a list in which a plurality of items are arranged in order of recommendation to a user; presenting, to the user, items included in the generated list in accordance with the arrangement order of the list; if an unnecessary item selection operation for designating an unnecessary item from among the presented items is made by the user, identifying a low-interest item similar to the unnecessary item and not yet presented to the user among the items arranged in the list; and lowering the order of the low-interest item in the list.

Description

Item list providing system, server, terminal, item list providing method, and program

[0001] The present invention relates to an item list providing system, a server, a terminal, an item list providing method, and a program.

[0002] A technology has been proposed for displaying a plurality of products related to a product in which a user has shown interest in a recommended product section in a shop or the like set up on the Internet (see, for example, Patent Document 1).

[0003] JP 2008-282098 A

[0004] When recommending items such as products to a user, it is desirable to reduce the possibility that an item that is expected to have little interest to the user will come to the user's attention.

[0005] The present invention has been made in view of the above-mentioned circumstances, and aims to provide an item list providing system and the like that can suitably recommend items to users.

[0006] In order to achieve the above object, the item list providing system of the present invention is an item list providing system comprising a server and a terminal, which generates a list in which a plurality of items are arranged in an order in which they are recommended to a user, presents the items included in the generated list to the user in the order in which they are arranged in the list, and when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifies from the items in the list low-interest items that are similar to the unnecessary items and have not yet been presented to the user, and the server and the terminal work together to perform a process of lowering the order of the low-interest items in the list.

[0007] The server of the present invention comprises: a generation unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a provision unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a reception unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed to designate an unnecessary item from the items displayed on the screen of the terminal; an identification unit that identifies low-interest items among the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low-interest items in the list.

[0008] The report may further include display information indicating which items from the list are displayed on the screen of the terminal, and items not shown in the display information may be designated as hidden items.

[0009] An item that has not yet been provided to the terminal may be the hidden item.

[0010] A new item that is not similar to the unnecessary item and is not included in the list may be newly recommended, and the new item may be added to the list between the ranking of the items already provided to the terminal and the ranking of the items of low interest.

[0011] The terminal of the present invention comprises: an acquisition unit that acquires from a server a list in which a plurality of items are arranged in the order in which they are recommended; a display unit that displays the acquired list in a scrollable manner on the screen; a reception unit that accepts an unnecessary selection operation that designates unnecessary items from the items arranged in the list and displayed on the screen; an identification unit that identifies low interest items among the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low interest items in the list.

[0012] Reporting the unwanted item to the server may cause the server to lower the priority of recommending the reported item.

[0013] By reporting the unwanted item to the server, the server may be caused to identify similar items that are similar to the unwanted item among the items in the list.

[0014] A table of similarities between items in the list may be obtained from the server along with the list, and the similarities between each of the hidden items and the unnecessary items may be found from the table.

[0015] A feature vector of each item in the list may be acquired from the server along with the list, and the similarity between each of the hidden items and the unnecessary items may be calculated based on the feature vector.

[0016] The similar items may be identified by comparing the determined similarity with a predetermined threshold, or a predetermined number or percentage of items with the highest determined similarity may be identified as the similar items and the items of least interest.

[0017] The server may be configured to newly recommend new items that are not similar to the unwanted items, and the new items may be added to the list in a higher order than the items of less interest.

[0018] The item list providing method of the present invention is an item list providing method in an item list providing system having a server and a terminal, which generates a list in which a plurality of items are arranged in an order in which they are recommended to a user, presents the items included in the generated list to the user in the order in which they are arranged in the list, and when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifies, from the items in the list, items of low interest that are similar to the unnecessary item and have not yet been presented to the user, and the server and the terminal work together to perform a process of lowering the order of the low interest item in the list.

[0019] The program of the present invention causes a computer to function as: a generation unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a provision unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a reception unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed on the terminal to designate an unnecessary item from the items displayed on the screen; an identification unit that identifies low-interest items among the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low-interest items in the list.

[0020] Another program of the present invention causes a computer to function as: an acquisition unit that acquires a list in which multiple items are arranged in the order in which they are recommended; a display unit that displays the acquired list in a scrollable manner on a screen; a reception unit that accepts an unnecessary selection operation to specify unnecessary items from the items arranged in the list and displayed on the screen; an identification unit that identifies low-interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low-interest items in the list.

[0021] According to the present invention, items can be suitably recommended to a user.

[0022] 1 is a block diagram showing the configuration of an item list providing system according to the present embodiment; FIG. 2 is a diagram showing an example of the hardware configuration of an information processing device that constitutes the item list providing system according to the present embodiment; FIG. 3 is a flowchart showing an example of a recommendation list presentation process according to embodiment 1; FIG. 4 is a flowchart showing an example of a recommendation list generation process; FIG. 5 is a diagram showing an example of a display of item information and a recommendation list; (A) to (D) are diagrams showing an example of a display of a recommendation list; (A) to (F) are diagrams showing an example of a recommendation list providing process according to embodiment 2; FIG. 6 is a diagram showing a method of managing a recommendation list according to embodiment 2; (A) to (F) are diagrams showing an example of updating the recommendation ranking of a recommendation list according to embodiment 2; FIG. 7 is a flowchart showing an example of a recommendation list display process according to embodiment 3; FIG. 8 is a diagram showing a method of managing a recommendation list according to embodiment 3; (A) to (F) are diagrams showing an example of updating the recommendation ranking of a recommendation list according to embodiment 3.

[0023] An item list providing system 100 according to an embodiment of the present invention will be described below with reference to the drawings. Note that the same or corresponding parts in the drawings are designated by the same reference numerals.

[0024] FIG. 1 is a block diagram showing the configuration of an item list providing system 100 according to this embodiment. The item list providing system 100 is a system that presents and provides a recommendation list, which is a list of recommended items, to a user. The item list providing system 100 is composed of a server 10 and a terminal 20. The server 10 and the terminal 20 are connected via a communication network 30 such as the Internet or a mobile communication network. The server 10 and the terminal 20 transmit and receive information via the communication network 30.

[0025] The server 10 stores multiple pieces of information about specific items, generates a recommendation list (recommended item list), which is a list in which items are recommended to the user in order of recommendation, and provides the list to the terminal 20 via a communication network. For example, the server 10 provides item information for a specified item in response to a request from the terminal 20 to browse a specific item specifying the item or a request to browse an item specifying search criteria. In this way, the server 10 also functions as a web server that provides a website whose content is item information. The server 10 then provides the terminal 20 with, along with the item information, a recommendation list in which items similar to the specified item are arranged in order of recommendation, for example. This allows the user of the terminal 20 to be provided with not only the item information for the specified item, but also a recommendation list of items recommended based on the specified item, thereby providing useful information to the user. Note that items recommended to the user using the recommendation list are sometimes referred to as recommended items.

[0026] In this embodiment, the server 10 provides the recommendation list in response to a request to view items from the terminal 20, but the recommendation list may also be provided to the user in response to other triggers, such as access to a specific website or a request to acquire the recommendation list itself specifying conditions, etc. Also, the server 10 may generate a recommendation list (a list of products, etc. recommended to the user) and provide it to the terminal 20 as an advertisement on a website or SNS (Social Networking Service).

[0027] An item may be any information that a user views, such as information about products, services, used cars, delivery services, digital content such as books, manga, music, and videos, accommodations, transportation (trains, buses, airplanes, etc.), travel, tours, restaurants, recipes, real estate, beauty facilities (hair salons, estheticians, etc.), golf courses, leisure facilities such as campgrounds and fishing boats, and sporting events, or information about purchasing or reserving any of these items.

[0028] The server 10 is configured by, for example, a communication-capable information processing device such as a personal computer (PC) or a workstation. The server 10 may also be a dedicated device. Although one server 10 is shown in FIG. 1, multiple servers 10 may be provided, for example, for each item to be stored and managed.

[0029] The terminal 20 is an information processing device with communication capabilities, such as a smartphone, tablet, or PC used by a user. The terminal 20 transmits an item viewing request to the server 10 via the communication network 30 based on a user's operation. The terminal 20 then acquires item information and a recommendation list corresponding to the viewing request, which are transmitted from the server 10 based on the item viewing request. The acquired item information is displayed on the screen of the terminal 20. The acquired recommendation list is also displayed in a scrollable manner on the screen of the terminal 20, allowing the user to sequentially view items in the list one by one. This allows the user to view the requested item information and also view the recommendation list, which is a list of items recommended based on the request. While FIG. 1 shows one terminal 20, any number of terminals 20 may be used.

[0030] 2 is a diagram showing an example of the hardware configuration of an information processing device that constitutes the server 10 and the terminal 20. The server 10 has a processor 11 such as a CPU (Central Processing Unit) or GPU (Graphical Processing Unit), a storage device 12 such as a memory, HDD and / or SSD, a communication interface 13 for wired or wireless communication, an input device 14 for accepting input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, a touch panel, a camera, a mouse, and / or a microphone. The output device 15 is, for example, a display and / or a speaker.

[0031] The storage device 12 of the server 10 stores information about multiple items under management. It also stores information (unnecessary item information, recommendation priority information) for each user who accesses the server 10 via the terminal 20. At least a portion of the information about the users may be stored in the storage device 22 of the terminal 20.

[0032] Similarly, the terminal 20 has a processor 21, a storage device 22, a communication interface 23, an input device 24, and an output device 25. Note that unnecessary components in the server 10 and the terminal 20 may be omitted as appropriate, or other components may be added.

[0033] Furthermore, the item list providing system 100 of this embodiment can accept from the user an unwanted selection operation to designate an unwanted item from among the items included in the recommendation list presented to the user. The item list providing system 100 then identifies low-interest items that are similar to the designated unwanted item and have not yet been presented to the user, and the server 10 and the terminal 20 cooperate to execute a process of lowering the recommendation rank of the low-interest items in the recommendation list. In this embodiment, the recommendation list can be presented sequentially by scrolling or the like, and the list is presented in parts. Lowering the recommendation rank (specifically, the display order of the list) of low-interest items that are similar to the unwanted item selected by the user and have not yet been presented to the user can reduce the user's visibility by scrolling or the like. This makes it possible to present a recommendation list that is suitable for the user's unwanted selection operation.

[0034] First Embodiment Hereinafter, first, a first embodiment will be described, which is an example in which the server 10 and the terminal 20 cooperate to manage a recommendation list in the entire item list providing system 100, and a process for lowering the recommendation ranking of a less interesting item in the recommendation list is executed. In the first embodiment, it is expressed that the item list providing system 100 "presents" the recommendation list.

[0035] 3 is a flowchart showing an example of a recommendation list presentation process executed by the item list providing system 100 of embodiment 1. The recommendation list presentation process is executed by the processor 11 of the server 10 and the processor 21 of the terminal 20 in cooperation with each other. Specifically, the process is executed by the processor 11 that has read a program stored in the storage device 12, and the processor 21 that has read a program stored in the storage device 22. Note that, among the processes in FIG. 3 , processes for which no particular subject is specified may be executed by at least either the server 10 or the terminal 20.

[0036] In the recommendation list presentation process, the item list providing system 100 first determines whether or not a user has made a request to view items related to the server 10 (step S11). A user's request to view items is, for example, an operation to view items using the terminal 20. If the user has not made a request to view items (step S11; No), the recommendation list presentation process ends.

[0037] If the user requests to view an item (step S11; Yes), the item list providing system 100 executes a recommendation list generation process to generate a recommendation list to be presented to the user who has requested to view the item (step S12).

[0038] FIG. 4 is a flowchart showing an example of a recommendation list generation process. The recommendation list generation process is executed by the processor 11 of the server 10. In the recommendation list generation process, the processor 11 first extracts items similar to the item requested by the user to view the item (step S121). In step S121, for example, a predetermined number of items having attribute information (category, manufacturer, condition, size, color, region, price range, etc.) similar to the item specified in the view request are extracted. Specifically, a list of multiple items (e.g., different colors, different manufacturers, etc.) belonging to the same category as the item requested to be viewed is created. Note that any algorithm may be used to extract similar items.

[0039] Next, the processor 11 acquires recommendation priority information of the user who has made the item viewing request (step S122).

[0040] Then, the processor 11 generates a recommendation list in which the items extracted in step S121 are arranged in the order in which they will be recommended to the user based on the recommendation priority information acquired in step S122 (step S123). Thereafter, the recommendation list generation process ends.

[0041] The recommendation priority information is information indicating the recommendation priority of items for each user, and is stored in the storage device 12 of the server 10 in association with the user's identification information, for example. When the user performs an unnecessary selection operation in the recommendation list presented to the user, the recommendation priority information is updated to lower the recommendation ranking of an item selected as an unnecessary item. The recommendation priority information may store at least information on unnecessary items selected as unnecessary. Furthermore, for example, when the user performs a browsing operation on a specific item, the recommendation priority information is updated to raise the priority of that item.

[0042] The recommendation priority information is referenced when a new recommendation list is generated in step S123. For example, the items extracted in step S121 are arranged in the recommendation list in descending order of priority. Furthermore, unnecessary items based on the user's past unnecessary selection operations may be excluded from the recommendation list, or the recommendation list may be generated with a lower recommendation order. This allows a recommendation list suited to the user to be generated.

[0043] It should be noted that the user's item browsing history may be stored together with the recommendation priority information, and the recommendation list may be generated by referring to the item browsing history.

[0044] Furthermore, for items for which there is no difference in recommendation priority based on the recommendation priority information, the recommendation order in the recommendation list may be determined randomly, or the recommendation order may be determined based on information about the item itself, such as the popularity of the item, how recent the information is, the number of items in stock, and other information, to generate the recommendation list. Furthermore, when referring to the user's item viewing history, the recommendation order may be increased for items that have been viewed in the past or have been viewed frequently.

[0045] Furthermore, for example, the recommendation list generation process may refer to trends in products that the user has viewed or purchased in the past, the user's own attributes (age, gender, preferences, etc.), and trends in products that other users who have attributes in common with the user's own attributes have viewed or purchased, and generate a list that is likely to be purchased or viewed based on, for example, machine learning.

[0046] 3, the item list providing system 100 presents the recommendation list generated in step S12 to the user together with item information corresponding to the browsing request (step S13). If the browsing request specifies a specific item, the item information corresponding to the browsing request may be information about the specific item (e.g., product overview, price information, etc.), or if the browsing request specifies search conditions for items, it may be search results for the items (a list of items matching the search conditions, overview, etc.).

[0047] FIG. 5 shows an example of a display of item information and a recommendation list presented by the item list providing system 100. FIG. 5 shows an example of a case where a user accesses an item search site using a web browser 110 installed on a terminal 20 and performs a browsing operation for a specific item, resulting in a web page being displayed on the web browser 110, including item information and a recommendation list for the specific item. As shown in FIG. 5 , the web browser 110 displays item information such as an item name 111 and detailed item information 112 in response to the item browsing operation (browsing request). A recommendation list 113 is also displayed along with the item information. The recommendation list 113 shown in FIG. 5 displays the top three items, i.e., items RI1, RI2, and RI3, as recommended items recommended to the user. Thus, in the display example shown in FIG. 5 , three items are displayed in each recommendation list generated by the item list providing system 100. For example, items RI1, RI2, and RI3 each display an overview of the item, and provide a link to detailed information about the item. Also, a scroll button 114 is displayed for scrolling through the items displayed in the recommendation list 113. The recommendation list 113 in FIG. 5 displays three items, but a recommendation list generated by the item list providing system 100 contains four or more items, and by operating the scroll button 114, the items are scrolled one by one in order according to the order of the item list (recommended ranking).

[0048] Next, the item list providing system 100 determines whether or not the user has performed an unnecessary selection operation specifying an unnecessary item in the recommendation list presented to the user (step S14).

[0049] If the user selects an unnecessary item (step S14; Yes), the item list providing system 100 sets the selected unnecessary item to not be displayed in the recommendation list (step S15). This prevents the user from continuing to see the unnecessary item. At this time, the selected unnecessary item is stored as an unnecessary item in the recommendation priority information corresponding to the user (for example, the terminal 20 reports the selected unnecessary item to the server 10, and the unnecessary item is stored in the recommendation priority information). Therefore, the unnecessary item will be less likely to be recommended to the user in the future, or will not be recommended at all.

[0050] An unnecessary selection button ds is displayed for each of the items RI1, RI2, and RI3 shown in FIG. 5 . Operating the unnecessary selection button ds selects the item as unnecessary, and the selected item is hidden. For example, FIG. 6 shows an example of a recommended list display. For example, as shown in FIGS. 5 and 6A , the top three recommended items RI1, RI2, and RI3 are displayed as default display items in the recommended list 113. When the unnecessary selection button ds for item RI2 is selected (unnecessary selection operation), item RI2 is hidden, as shown in FIG. 6B , and a deleted item DI is displayed in the place where item RI2 was displayed. Note that an undo button rs is displayed for the deleted item DI. Operating this undo button rs cancels the unnecessary selection operation, and item RI2 is redisplayed. This allows the operation to be canceled if the unnecessary selection button ds is operated by mistake. In this case, subsequent processes such as identifying items of low interest are not executed. The deleted item DI may be displayed in a manner that allows the item before deletion to be identified. For example, the item RI2 before deletion may be displayed in a dim or dark color in the background of the deleted item DI.

[0051] Then, the item list providing system 100 executes a low interest item identification process to identify low interest items that are similar to the unwanted items selected by the user as unwanted and have not been presented to the user from the items included in the recommendation list (step S16).

[0052] In the low interest item identification process, for example, an unnecessary item is compared with other items included in the recommendation list that have not been presented to the user, and items similar to the unnecessary item included in the recommendation list are identified as low interest items.

[0053] Similar items may be determined, for example, based on the degree of commonality of attribute information (category, manufacturer, condition, size, color, region, price range, etc.), or based on a similarity table prepared in advance that classifies the similarity of each item. Also, feature vectors or feature amounts of elements of each item (image data, text data, etc.) may be calculated, and similar items may be determined based on the similarity of the feature vectors or feature amounts. In other words, similar items may be identified using any method.

[0054] Alternatively, the items included in the recommendation list may be compared with unnecessary items to determine their similarity, and items whose similarity is within a predetermined threshold may be extracted as similar items (items of low interest).Alternatively, a predetermined number or a predetermined percentage of items with the highest similarity may be extracted as similar items (items of low interest).

[0055] Whether or not an item has not yet been presented in the recommendation list can be determined by acquiring the browsing status by scrolling the recommendation list on the terminal 20, or by determining whether or not the item has already been provided to the terminal 20 from the server 10, etc.

[0056] When a low-interest item is identified, the item list providing system 100 lowers the recommendation rank of the low-interest item in the recommendation list (step S18). In this way, the recommendation rank of a low-interest item that has not yet been presented to the user is lowered in the recommendation list, so that when the user sequentially browses the recommended items, it is possible to reduce the possibility that an item that is expected to have little interest to the user will come into view.

[0057] The item list providing system 100 then updates the recommendation list by adding newly recommended items according to the number of low-interest items in the recommendation list (step S19). For example, the recommendation order of the low-interest items in the recommendation list is lowered to the lowest, and new recommended items, the same number as the low-interest items, are added to the recommendation list at a higher order than the low-interest items. In other words, the item list providing system 100 adds the newly recommended items between the order of the items already presented and the order of the low-interest items. This allows new items to be added to the recommendation list according to the number of low-interest items, and the new items can be recommended at a higher order than the low-interest items. Note that the method for extracting the recommended items to be added may use the same algorithm as when generating the recommendation list. For example, the recommendation list may be added with recommended items that are not included in the initially generated recommendation list, are the next most recommended items, and do not correspond to low-interest items.

[0058] The number of recommended items to be replenished is not limited to the same number as the items of low interest, but may be a predetermined ratio of the number of items of low interest, or may be more or less than the items of low interest. Furthermore, the ranking of new recommended items when replenishing may be arbitrary.

[0059] After the process of step S19 or if the user has not performed an unnecessary selection operation (step S14; No), the item list providing system 100 determines whether the user has performed a scroll operation on the recommendation list (step S20).

[0060] When the user performs a scroll operation (step S20; Yes), the item list providing system 100 scrolls and displays the items in the recommendation list in accordance with the recommendation order of the recommendation list, depending on the content of the scroll operation (step S21).

[0061] For example, as shown in FIG. 6(B), when two items RI1 and RI3 and a deleted item DI are displayed in the recommendation list 113, if the scroll button 114 is operated twice, the recommendation order in the recommendation list 113 shifts two levels, and items RI3, RI4, and RI5 are displayed, as shown in FIG. 6(C). Furthermore, if the scroll button 114 is operated three times, the recommendation order in the recommendation list 113 shifts three levels, and items RI6, RI8, and RI9 are displayed, as shown in FIG. 6(D). At this time, for example, item 7 corresponding to item RI7 is determined to be a low-interest item similar to item 2 corresponding to item RI2 selected as unnecessary, and its recommendation order is lowered to the bottom. Therefore, item RI7, which is an overview of item 7, is not displayed in FIG. 6(D). Furthermore, when the recommendation list 113 is scrolled from the default state, a scroll button 115 for scrolling the recommendation list 113 in the reverse direction is displayed, as shown in FIGS. 6(C) and 6(D).

[0062] In this embodiment, the recommendation list is scrolled in response to the operation of the scroll buttons 114 and 115, but the recommendation list may be automatically scrolled (updated) in response to a viewing operation of the recommendation list, such as a screen scroll operation. Also, in the examples shown in Figures 5 and 6, the recommendation list is scrolled left and right, but it may also be scrolled up and down, or up and down, left and right.

[0063] After the process of step S21 or if the user has not performed a scrolling operation on the recommendation list (step S20; No), the item list providing system 100 determines whether the user has performed another operation on the presented item information and recommendation list (step S22). If the user has performed another operation (step S22; Yes), control corresponding to the operation (for example, control of displaying item information of the selected recommended item) is executed (step S23). If the user has not performed another operation (step S22; No), or after the process of step S23, the process returns to step S14.

[0064] As described above, according to the item list providing system 100 of embodiment 1, the item list providing system 100 includes a server 10 and a terminal 20, generates a list in which a plurality of items are arranged in the order in which they are recommended to the user, and presents the items included in the generated list to the user according to the order in which they are recommended in the recommendation list. When the user performs an unnecessary selection operation to specify an unnecessary item from the presented items, the server 10 and the terminal 20 work together to identify low-interest items from the items in the recommendation list that are similar to the unnecessary items and have not yet been presented to the user, and then lower the recommendation order of the low-interest items in the recommendation list.

[0065] This lowers the recommendation rank (display order) of items in the recommendation list that are expected to have little interest to the user based on the unnecessary items selected by the user, thereby reducing the possibility that items that are expected to have little interest to the user will come into view as the user sequentially browses the recommended items.

[0066] In addition, since the recommendation ranking of low-interest items is lowered, other items are displayed preferentially in the recommendation list. As a result, other items are relatively more likely to be seen by the user, and a recommendation list containing items that are expected to be of high interest to the user can be presented.

[0067] Furthermore, the recommendation order in the recommendation list is changed within the range of items not yet presented to the user, so the order in the recommendation list can be changed without the user noticing.

[0068] Second Embodiment Next, a second embodiment will be described, which is an example in which the server 10 in the item list providing system 100 manages the recommendation list and executes a process of lowering the recommendation ranking of an item of low interest in the recommendation list. In the second embodiment, it is expressed that the server 10 “provides” the recommendation list to the terminal 20.

[0069] The server 10 of the second embodiment generates a recommendation list in which a plurality of items are arranged in the order in which they are recommended to the user. The server 10 provides the items included in the generated list to the terminal 20 according to the order in which they are arranged in the recommendation list, and displays them on the screen of the terminal 20. The server 10 receives a report from the terminal 20 indicating that an unnecessary selection operation has been performed to designate an unnecessary item from the items displayed on the screen of the terminal 20. The server 10 then identifies, from among the items arranged in the recommended items, low-interest items that are similar to the unnecessary item and are hidden items that have not yet been displayed on the screen of the terminal 20. Furthermore, the server 10 updates the recommendation list by lowering the display priority of low-interest items in the recommendation list.

[0070] 7 is a flowchart showing an example of a recommendation list providing process executed by the server 10 according to embodiment 2. The recommendation list providing process is executed by the processor 11 of the server 10 reading a program stored in the storage device 12.

[0071] In the recommendation list providing process, the server 10 first determines whether or not a user request to view an item related to the server 10 has been received from the terminal 20 (step S51). The user request to view an item is transmitted, for example, in response to an item viewing operation on the terminal 20. If the user request to view an item has not been received (step S51; No), the recommendation list providing process is terminated.

[0072] If an item browsing request has been received (step S51; Yes), the server 10 executes a recommendation list generation process for generating a recommendation list to be provided to the terminal 20 of the user who made the item browsing request (step S52). The recommendation list generation process may be the same as the process of the first embodiment shown in Fig. 4. The process of step S52 corresponds to the generation unit of the server 10.

[0073] Upon completion of the recommendation list generation process, the server 10 provides the recommendation list generated in step S52 together with the item information corresponding to the browsing request to the user's terminal 20 (step S53). The server 10 transmits, for example, web page data including the item information and the recommendation list, as shown in FIG. 5, to the terminal 20. Note that the content of the item information corresponding to the browsing request may be information about a specific item or a search result for an item, as in the first embodiment. The processing of step S53 corresponds to the providing unit of the server 10.

[0074] In the second embodiment, the server 10 provides the terminal 20 with a portion of the recommended items included in the recommendation list generated by the server 10. In the recommendation list generation process in step S52, the server 10 generates a recommendation list including a predetermined number of items. The predetermined number may be any number and may vary depending on the number of similar items extracted in the recommendation list generation process. The server 10 then provides the terminal 20 with a portion of the recommended items included in the generated recommendation list. The number of portions provided may be any number as long as it is equal to or less than the predetermined number. It may also be a predetermined percentage (e.g., 30%) of the predetermined number. Then, when the recommendation list is scrolled on the terminal 20 and all of the provided items have been viewed, the server 10 provides the next item from the recommendation list.

[0075] Fig. 8 is a diagram showing a method for managing a recommendation list in the server 10. As shown in Fig. 8, the server 10 generates a recommendation list in response to a request to view items from the terminal 20. The recommendation list shown in Fig. 8 includes 12 items. First, the server 10 manages the status of all items included in the generated recommendation list as "not provided" (not provided to the terminal 20) (Fig. 8 (1)).

[0076] The server 10 then provides some of the items included in this recommendation list to the terminal 20. In Fig. 8, items 1 to 5, which are the top five items recommended in the recommendation list, are provided to the terminal 20. The server 10 updates the status of these items to "Provided" (provided to the terminal 20) (Fig. 8 (2)).

[0077] Next, the server 10 determines whether or not it has received from the terminal 20 unnecessary selection operation information indicating that the user has performed an unnecessary selection operation specifying an unnecessary item in the recommendation list provided to the user (step S54). The unnecessary selection operation information is a report from the terminal 20 indicating that an unnecessary selection operation has been performed. The processing of step S54 corresponds to the receiving unit of the server 10.

[0078] When the server 10 receives the unnecessary selection operation information from the terminal 20 (step S54; Yes), the server 10 sets the selected unnecessary item to be hidden in the recommendation list (step S55). For example, the server 10 transmits to the terminal 20 data of a web page in which the portion where the unnecessary item was displayed has been updated to be hidden (for example, the deleted item DI shown in FIG. 6B), and updates the recommendation list displayed on the terminal 20.

[0079] This prevents the user from continuously seeing items that are not needed by the user. At this time, the item selected as unnecessary is stored in the storage device 12 of the server 10. Specifically, it is stored and updated as an unnecessary item in the recommendation priority information according to the user. Therefore, the unnecessary item will be less likely to be recommended to the user in the future, or will not be recommended at all.

[0080] The server 10 then executes a low interest item identification process to identify low interest items that are similar to the unwanted item selected by the user and have not yet been provided to the user from the items included in the recommendation list (step S56). The process of step S56 corresponds to the identification unit of the server 10.

[0081] In the low interest item identification process, for example, the server 10 compares the unnecessary item with other items included in the recommendation list, and identifies items similar to the unnecessary item included in the recommendation list.

[0082] Similar items may be determined, for example, based on the degree of commonality of attribute information (category, manufacturer, condition, size, color, region, price range, etc.), or based on a similarity table prepared in advance that classifies the similarity of each item. Also, feature vectors or feature amounts of elements of each item (image data, text data, etc.) may be calculated, and similar items may be determined based on the similarity of the feature vectors or feature amounts. In other words, similar items may be identified using any method.

[0083] The server 10 may also compare items included in the recommendation list with unnecessary items to determine similarities, and extract items whose similarities are within a predetermined threshold as similar items (items of low interest).Also, the server 10 may extract a predetermined number or a predetermined percentage of items with the highest determined similarities as similar items (items of low interest).

[0084] Items that have not been provided from the server 10 to the terminal 20 are not displayed on the terminal 20 and have not yet been viewed by the user. Therefore, in step S56, the items that have not been provided are approximated to items that have not been displayed, and items of low interest are identified from the items that have not been provided. This eliminates the need for the server 10 to obtain the viewing status of recommended items from the terminal 20, thereby reducing the processing load.

[0085] On the other hand, it is also possible that the number of recommended items provided from the server 10 to the terminal 20 at one time is not the same as the number of recommended items displayed on the terminal 20 at one time. For example, if the number of recommended items that can be displayed on the terminal 20 at one time is smaller than the number of recommended items provided from the server 10 to the terminal 20 at one time, there will be items that have been provided but not yet displayed. In this case, display information indicating whether or not an item has actually been displayed on the terminal 20 may be obtained, and low-interest items may be identified from the undisplayed items based on the display information. In this case, if the items that have been provided to the terminal 20 but not yet displayed are low-interest items and the recommendation list for that portion is updated, the updated recommendation list may be transmitted to the terminal 20.

[0086] When a low-interest item is identified, the server 10 lowers the recommendation rank of the low-interest item in the recommendation list (step S58).The server 10 then updates the recommendation list by adding newly recommended items according to the number of low-interest items in the recommendation list (step S59).The server 10 adds the newly recommended item between the rank of the previously presented item and the rank of the low-interest item.The processes of steps S58 and S59 are the same as steps S18 and S19 in embodiment 1.The processes of steps S58 and S59 correspond to the update unit of the server 10.

[0087] For example, as shown in FIG. 8 (2), when the server 10 is providing the terminal 20 with items 1 through 5 included in the recommendation list, if the terminal 20 transmits information about the third item as an unnecessary selection operation to the server 10, items similar to the third item from items 6 through 12 that have not yet been provided in the recommendation list are identified as low-interest items. The identified low-interest items are then lowered to the bottom of the recommendation ranking. At this time, the server 10 adds the same number of recommended items as the low-interest items, but at a higher ranking (e.g., one rank higher than the low-interest items) than the low-interest items. In this way, the server 10 can update the recommendation list in response to the user's unnecessary selection operation, thereby sequentially providing the user with a recommended list that is suitable for the user. The server 10 also updates the status of the third item selected as unnecessary to "deleted" and the status of the low-interest items to "low-interest items" (FIG. 8 (3)).

[0088] After processing of step S59 or if unnecessary selection operation information has not been received from the terminal 20 (step S54; No), the server 10 determines whether or not scrolling operation information indicating that the user has performed a scrolling operation on the recommendation list has been received from the terminal 20 (step S60).

[0089] When the server 10 receives the scroll operation information from the terminal 20 (step S60; Yes), the server 10 scrolls the items in the recommendation list displayed on the screen of the terminal 20 in accordance with the recommendation order of the recommendation list in accordance with the content of the scroll operation information (step S61). For example, the server 10 transmits data of the web page updated by scrolling the recommendation list in accordance with the recommendation order to the terminal 20, thereby updating the recommendation list displayed on the terminal 20.

[0090] For example, as shown in FIG. 8(3), if the server 10 receives scroll operation information from the terminal 20 regarding items 1 to 5 included in the recommendation list from the server 10 to the terminal 20, the server 10 will provide the terminal 20 with items 6 to 10, which are ranked sixth to tenth in the recommendation order. In this way, the server 10 can sequentially provide recommended items to the terminal 20. The server 10 updates the status of these newly provided items to "provided" (provided to the terminal 20) (FIG. 8(4)).

[0091] After the processing of step S61 or when the server 10 has not received information on the scroll operation of the recommendation list from the terminal 20 (step S60; No), the server 10 determines whether or not other operation information indicating that the user has performed another operation on the provided item information and recommendation list has been received from the terminal 20 (step S62). If other operation information has been received from the terminal 20 (step S62; Yes), the server 10 executes control according to the operation information (for example, display control of the item information of the selected recommended item, web page update control, etc.) (step S63). If other operation information has not been received from the terminal 20 (step S62; No) or after the processing of step S63, the server 10 returns to the processing of step S54.

[0092] 9 is a diagram showing an example of updating the recommendation ranking of a recommendation list generated by the server 10. For example, a description will be given of a case where a recommendation list consisting of items 1 to 12 is generated by the server 10. The server 10 provides, for example, five recommended items. Immediately after generating the recommendation list, items 1 to 5 have been provided to the terminal 20, as shown in FIG. 9A. When three items are displayed at a time on the terminal 20, items 1 to 3 have been displayed to the user, and items 4 and 5 have not yet been displayed, as shown in FIG. 9A.

[0093] Thereafter, for example, when the user scrolls the recommendation list by one step, item 4 is displayed as shown in FIG. 9B.

[0094] Here, for example, if the user selects item 3 as unnecessary, item 3 is deleted and becomes an unnecessary item, as shown in FIG. 9C.

[0095] Then, from items 6 to 12, which are unprovided items, items similar to the unwanted item selected as unwanted are identified, and for example, as shown in Figure 9 (D), the identified item 7 is extracted as an item of low interest.

[0096] Thereafter, as shown in FIG. 9(E), items 13, the same number as the extracted low interest items, are added as recommended items in the order just before the low interest items, and as shown in FIG. 9(F), item 7, which is a low interest item, is moved to the bottom of the recommendation list.

[0097] In this way, the recommendation list is dynamically updated in the server 10 in response to operations on the recommendation list on the terminal 20. The updated recommendation list is supplied to the terminal 20 by an operation to acquire a new item, such as a scrolling operation on the terminal 20. This allows recommended items to be suitably supplied to the user when the user sequentially browses the recommended items. Specifically, it is possible to reduce the possibility that an item that is expected to be of little interest will come to the user's eye.

[0098] As described above, according to the server 10 of embodiment 2, a recommendation list is generated in which multiple items are arranged in the order in which they are recommended to the user, the items included in the generated recommendation list are provided to the terminal 20 according to the recommendation order in the recommendation list and displayed on the screen of the terminal 20, a report is received from the terminal indicating that an unnecessary selection operation has been performed to designate an unnecessary item from the items displayed on the screen of the terminal 20, low-interest items are identified from the items arranged in the recommendation list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen, and the order of the low-interest items is lowered in the recommendation list.

[0099] As a result, based on the unnecessary items selected by the user, the server 10 can take the initiative to lower the recommendation rank (display order) in the recommendation list of items that are expected to have little interest to the user, and provide recommended items from that recommendation list to the terminal 20. Therefore, when the user sequentially browses the recommended items, it is possible to reduce the possibility that items that are expected to have little interest to the user will come into view.

[0100] In addition, since the recommendation ranking of low-interest items is lowered, other items are displayed preferentially in the recommendation list, which makes it relatively easier for the user to see other items, making it possible to provide a recommendation list filled with items that are expected to be of high interest to the user.

[0101] Furthermore, the recommendation order in the recommendation list is changed within the range of items not yet provided to the user, so the order in the recommendation list can be changed without the user noticing.

[0102] (Embodiment 3) Next, a description will be given of embodiment 3, which is an example of a case where the terminal 20 in the item list providing system 100 manages the recommendation list and executes a process of lowering the recommendation ranking of an item of low interest in the recommendation list. In embodiment 2, it is expressed that the terminal 20 "displays" the recommendation list on the screen.

[0103] The terminal 20 of the third embodiment acquires, in a lump, from the server 10, a recommendation list in which a plurality of items generated by the server 10 are arranged in the order in which they are recommended. The terminal 20 displays the acquired list on the screen in a scrollable manner. The terminal 20 accepts an unnecessary selection operation to designate unnecessary items from the items arranged in the recommendation list and displayed on the screen. Among the items arranged in the recommendation list, low-interest items are identified, which are similar items to the unnecessary items and are hidden items not yet displayed on the screen. The terminal 20 updates the recommendation list by lowering the order of the low-interest items in the recommendation list.

[0104] 10 is a flowchart showing an example of a recommendation list display process executed by the terminal 20 according to embodiment 3. The recommendation list display process is executed by the processor 21 of the terminal 20 reading a program stored in the storage device 22.

[0105] In the recommendation list display process, the terminal 20 first determines whether or not a user's item browsing operation has been detected (step S71). User's item browsing operations include, for example, browsing items provided by the server 10, accessing a search website, selecting item information on the website, searching, etc. If an item browsing operation has not been detected (step S71; No), the recommendation list display process is terminated.

[0106] If an item browsing operation is detected (Step S71; Yes), the terminal 20 transmits an item browsing request corresponding to the user's browsing operation to the server (Step S72). The terminal 20 then waits until it receives item information corresponding to the item browsing request and an accompanying recommendation list from the server 10 (Step S73). The recommendation list is generated by the server 10 executing the recommendation list generation process shown in FIG. 4, for example. The process of Step S73 corresponds to the acquisition unit of the terminal 20.

[0107] When the item information and the recommendation list are received from the server (step S73; Yes), the terminal 20 displays the received item information and the recommendation list on the screen (step S74). The terminal 20 displays, for example, web page data including the item information and the recommendation list, as shown in Fig. 5. Note that the content of the item information may be information about a specific item or a search result for an item, as in the first embodiment. The processing of step S74 corresponds to the display unit of the terminal 20.

[0108] In the third embodiment, the terminal 20 is configured to display, one by one, the recommended items included in the recommendation list received in a lump sum from the server 10. The server 10 generates a recommendation list including a predetermined number of items, which is an arbitrary number. The terminal 20 then displays, one by one, the items included in the received recommendation list on the terminal 20. The number of items to be displayed one by one may be any number as long as it is equal to or less than the predetermined number. It may also be a predetermined percentage (e.g., 30%) of the predetermined number. Then, when the recommendation list is scrolled on the terminal 20 and all of the displayed items have been viewed, the terminal 20 displays the next item from the recommendation list.

[0109] FIG. 11 is a diagram illustrating a method for managing a recommendation list on the terminal 20. As shown in FIG. 11 , the terminal 20 transmits an item viewing request to the server 10 in response to a user's item viewing operation. In response to the viewing request, the server 10 generates a recommendation list and transmits the generated recommendation list to the terminal 20 in a batch. At this time, the server 10 determines the similarity between the recommended items included in the recommendation list and also transmits the resultant similarity information between the recommended items. As in the above embodiment, the method for determining the similarity between the recommended items may be based on the similarity of attribute information, feature vectors, feature quantities, etc. Therefore, the similarity information may be a similarity table that classifies the similarity between each item, or information representing the feature vectors of each item. By transmitting the similarity information in advance in this manner, the terminal 20 does not need to calculate the similarity between the recommended items, thereby reducing the processing load on the terminal 20 and speeding up processing.

[0110] The recommendation list transmitted from the server 10 shown in Fig. 11 includes 12 items. The terminal 20 displays a portion of the items included in this recommendation list (for example, three items at a time) on the terminal 20. In Fig. 11, items 1 to 3, which are the top three items recommended in the recommendation list, are displayed. The terminal 20 manages the status of the displayed items as "displayed" (displayed on the terminal 20) and the status of the other items as "not displayed" (not displayed on the terminal 20) (Fig. 11 (1)).

[0111] Next, the terminal 20 determines whether or not it has received a user's selection operation for specifying an unnecessary item in the displayed recommendation list (step S75). The processing of step S75 corresponds to a receiving unit of the terminal 20.

[0112] If the unnecessary selection operation is received (step S75; Yes), the terminal 20 sets the selected unnecessary item to hidden in the recommendation list (step S76). For example, the terminal 20 updates the recommendation list by hiding the portion where the unnecessary item was displayed (for example, the deleted item DI shown in FIG. 6(B)). This prevents the user from continuing to see the unnecessary item. At this time, the item selected as unnecessary is stored in the storage device 22 of the terminal 20. Specifically, the item is stored and updated as an unnecessary item in the recommendation priority information corresponding to the user. Therefore, the unnecessary item will be less likely to be recommended to the user in the future, or will not be recommended at all.

[0113] Then, the terminal 20 executes a low interest item identification process to identify low interest items that are similar to the unnecessary item selected by the user and have not yet been displayed to the user from the items included in the recommendation list (step S77). The process of step S77 corresponds to the identification unit of the terminal 20.

[0114] In the low interest item identification process, for example, the terminal 20 compares the unnecessary item with other items included in the recommendation list that have not been displayed to the user, and identifies items similar to the unnecessary item included in the recommendation list as low interest items.

[0115] Similar items to the unwanted item may be identified based on similarity information (similarity table, feature vector, etc.) received from the server 10 along with the recommendation list. For example, based on the similarity information, the items included in the recommendation list may be compared with the unwanted item to determine the similarity, and items whose similarity is within a predetermined threshold may be extracted as similar items (items of low interest). Also, a predetermined number or a predetermined percentage of items with the highest determined similarity may be extracted as similar items (items of low interest).

[0116] When the terminal 20 detects an unnecessary selection operation, the terminal 20 may transmit information about the selected unnecessary item to the server 10, causing the server 10 to identify similar items and items of low interest. The method for determining similar items in the server may be, for example, the method described in embodiment 2. In this case, it is not necessary to obtain similarity information from the server 10 in advance.

[0117] Furthermore, if similarity information is not received in advance, the terminal 20 may acquire similarity information between recommended items when it detects an unnecessary selection operation. Also, the similarity table may be acquired at a timing other than when the recommendation list is acquired.

[0118] Items that are not displayed on the terminal 20 are items that have not yet been viewed by the user. In this embodiment, in order to change the recommendation order of such recommendation lists that have not yet been viewed by the user, in step S77, items of low interest are identified from the undisplayed items.

[0119] When the terminal 20 identifies the low-interest item, the terminal 20 lowers the recommendation rank of the low-interest item in the recommendation list (step S79). The process of step S79 is the same as step S18 in the first embodiment. Step S79 corresponds to the update unit of the terminal 20.

[0120] The terminal 20 then transmits to the server 10 unwanted item information, which is information about the items selected as unnecessary, and an item replenishment request, which requests that newly recommended items be replenished according to the number of low-interest items in the recommendation list (step S80). By transmitting the unwanted item information to the server 10, the server 10 can update the recommendation priority information for each user based on the unwanted item information. The unwanted selection operation information is a report of unwanted items from the terminal 20 to the server 10. The recommendation priority information is referenced when the server 10 generates a new recommendation list, and the recommendation list is generated by lowering the recommendation order of unwanted items based on the user's past unnecessary selection operation, so that a recommendation list suitable for the user can be generated. In other words, the report of unwanted items from the terminal 20 can cause the server 10 to lower the priority of recommending the reported items.

[0121] Thereafter, the terminal 20 waits until it receives replenishment item information in response to the item replenishment request from the server 10 (step S81). The item replenishment request includes, for example, information that can identify the number of recommendation lists and low-interest items.

[0122] When the server 10 receives an item replenishment request from the terminal 20, it extracts recommended items to be newly added to the recommendation list and transmits item information of the extracted recommended items to the terminal 20. For example, the server 10 may extract the same number of new recommended items as the low-interest items. The method for extracting the items to be replenished may use the same algorithm as used when generating the recommendation list. For example, the server 10 may replenish the recommendation list with recommended items that are not included in the initially generated recommendation list, are next most recommended items, and do not correspond to low-interest items.

[0123] When replenishment item information corresponding to the item replenishment request is received from the server 10 (step S81; Yes), the terminal 20 updates the recommendation list by replenishing the items corresponding to the received replenishment item information as newly recommended items (step S82). In step S82, for example, the recommendation order of low-interest items is lowered to the bottom in the recommendation list, and the new recommended items are added to the recommendation list at a higher order than the low-interest items. In other words, the terminal 20 adds the newly recommended items between the order of the items already presented and the order of the low-interest items. This allows new items to be replenished in the recommendation list according to the number of low-interest items, and the new items can be recommended at a higher order than the low-interest items.

[0124] After the process of step S82 or when the unnecessary selection operation has not been received (step S75; No), the terminal 20 determines whether or not it has detected that the user has performed a scroll operation on the recommendation list (step S83).

[0125] If a scroll operation is detected (step S83; Yes), the terminal 20 scrolls and displays the items in the recommendation list displayed on the screen of the terminal 20 in accordance with the recommendation order of the recommendation list in accordance with the content of the scroll operation (step S84). For example, the terminal 20 scrolls the recommendation list in accordance with the recommendation order to display updated web page data, thereby updating the recommendation list displayed on the terminal 20.

[0126] After the process of step S84 or if a scrolling operation of the recommendation list has not been detected (step S83; No), the terminal 20 determines whether or not it has detected that the user has performed another operation on the displayed item information and recommendation list (step S85). If another operation has been detected (step S85; Yes), the terminal 20 executes control according to the detected operation (for example, display control of item information of the selected recommended item, web page update control, etc.) (step S86). If another operation has not been detected (step S85; No), or after the process of step S86, the process returns to step S75.

[0127] In the third embodiment, as shown in Fig. 11(1), when items 1 to 3 included in the recommendation list are displayed on the terminal 20, if a scroll operation is performed on the terminal 20 to view, for example, the fourth item, the terminal 20 updates the status of the displayed item to "displayed" (displayed on the terminal 20) (Fig. 11(2)). Note that in this embodiment, as shown in Fig. 5, since the items in the recommendation list are displayed in groups of three, items 2 to 4 are displayed on the terminal 20 after the scroll operation.

[0128] Also, for example, as shown in Figure 11 (2), when items 2 to 4 included in the recommendation list are displayed on terminal 20, if an operation to select the third item as unnecessary is performed, the third item is hidden on terminal 20, and the status of the third item selected as unnecessary is updated to "deleted."

[0129] When an unnecessary item selection operation is performed on the terminal 20, unnecessary item information and an item replenishment request are sent to the server 10. Furthermore, the terminal 20 identifies an item similar to the third item from the fifth through twelfth items not yet displayed in the recommendation list as an item of low interest. The terminal 20 updates the status of the identified low-interest item to "low-interest item" (FIG. 11 (4)). In FIG. 11 (4), the fifth and eighth items are identified as items of low interest.

[0130] Then, the terminal 20 lowers the recommendation ranking of the identified low-interest item to the lowest, and updates the recommendation list by replenishing the items (here, two items) corresponding to the replenishment item information received from the server 10 at a higher ranking than the low-interest item (for example, one rank higher than the low-interest item) (FIGS. 11 (5) and (6)). In this way, in the third embodiment, the terminal 20 can update the recommendation list in response to the user's unnecessary selection operation, and can sequentially display recommendation lists suitable for the user.

[0131] 12 is a diagram showing an example of updating the recommendation ranking of a recommendation list on terminal 20. For example, a description will be given of a case where a recommendation list consisting of items 1 to 12 is generated on terminal 20. Terminal 20 displays, for example, three recommended items at a time. Immediately after receiving the recommendation list all at once from server 10, items 1 to 3 are already displayed on terminal 20, and items 4 to 12 are not yet displayed, as shown in FIG. 12(A).

[0132] Thereafter, for example, when the user scrolls the recommendation list by one step, item 4 is displayed as shown in FIG. 12(B).

[0133] Here, for example, if the user selects item 3 as unnecessary, item 3 is deleted and becomes an unnecessary item, as shown in FIG. 12(C).

[0134] Then, from items 5 to 12, which are undisplayed items, items similar to the unnecessary item selected as unnecessary are identified, and for example, as shown in Figure 12 (D), the identified items 5 and 7 are extracted as items of low interest.

[0135] Then, as shown in Figure 12 (E), items 13 and 14, the same number as the extracted low interest items, are added from server 10 as recommended items in the order just before the low interest items, and as shown in Figure 12 (F), items 5 and 7, which are low interest items, are moved to the bottom of the recommendation list.

[0136] In this way, the recommendation list is dynamically updated on the terminal 20 in response to operations on the recommendation list on the terminal 20. The updated recommendation list is sequentially displayed by operations to display new items, such as a scroll operation on the terminal 20. This allows recommended items to be appropriately provided to the user when the user sequentially browses the recommended items. Specifically, it is possible to reduce the possibility that items that are expected to be of little interest will come to the user's attention.

[0137] As described above, according to the terminal 20 of embodiment 3, a recommendation list in which multiple items are arranged in the order in which they are recommended is acquired in one go from the server 10, the acquired recommendation list is displayed on the screen in a scrollable manner, an unnecessary selection operation is accepted to specify unnecessary items from the items arranged in the recommendation list and displayed on the screen, low interest items are identified from the items arranged in the recommendation list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen, and the ranking of the low interest items is updated by lowering the ranking in the list.

[0138] As a result, based on the unnecessary items selected by the user, the terminal 20 can take the initiative in lowering the recommendation rank (display order) of items in the recommendation list that are expected to have little interest to the user, and display recommended items from that recommendation list on the terminal 20. Therefore, when the user sequentially browses the recommended items, it is possible to reduce the possibility that items that are expected to have little interest to the user will come into view.

[0139] In addition, since the recommendation ranking of low-interest items is lowered, other items are displayed preferentially in the recommendation list, which makes it relatively easier for the user to see other items, making it possible to provide a recommendation list filled with items that are expected to be of high interest to the user.

[0140] Furthermore, the recommendation order in the recommendation list is changed within the range of items not yet displayed to the user, so the order in the recommendation list can be changed without the user noticing.

[0141] (Modifications) The present invention is not limited to the above-described embodiment, and various modifications and applications are possible. For example, some of the above-described embodiments may be omitted or replaced, or any configuration may be added.

[0142] Instead of changing the order in which low-interest items are recommended in the recommendation list, the display format of the low-interest items may be changed. For example, the display format of the recommendation list may be changed so that the low-interest items are less noticeable. For example, the display size of the low-interest items in the list may be reduced or they may be removed from the recommendation list. This also reduces the possibility that items that are expected to be of low interest to the user will be noticed by the user as the user sequentially browses the recommended items.

[0143] When a browsing operation of an item included in the recommendation list (for example, touching or clicking a link to detailed information) is performed, it is inferred that the user has a high interest in that item, so that in the recommendation list, unpresented (undisplayed) highly-interested items that are similar to that item may be identified, and the recommendation ranking of the highly-interested items in the recommendation list may be raised. In this way, items of high interest in the recommendation list are more likely to be seen by the user, and items of relatively low interest are less likely to be seen by the user, so that items can be appropriately recommended to the user.

[0144] In the above embodiments, when an unnecessary item is selected, a new recommended item is added when the recommended rank of the unnecessary item is lowered. However, this is not limited to this, and the recommended rank of the unnecessary item may be lowered without adding a new recommended item. That is, the process of adding a new recommended item in step S19 of FIG. 3, step S59 of FIG. 7, and step S82 of FIG. 10 may be omitted.

[0145] The process of adding new recommended items to the recommendation list may be performed at different times. For example, new recommended items may be added to the recommendation list when a user scrolls the recommendation list. In this way, recommended items can be added sequentially as the user sequentially browses the recommendation list, so that a never-ending supply of recommended items can be provided.

[0146] The server 10 and the terminal 20 can be realized using a normal computer, not a dedicated device. For example, the server 10 and the terminal 20 that execute the above-described processes may be configured by installing a program for executing the above-described functions on a computer from a recording medium that stores the program. Also, a single server 10 may be configured by multiple computers operating in cooperation with each other.

[0147] Furthermore, the method for supplying the program to the computer is arbitrary. For example, the program may be supplied via a communication line, a communication network, a communication system, etc. For example, a non-transitory computer-readable recording medium storing the program for executing the above-described functions may be distributed.

[0148] Furthermore, when an OS (Operating System) provides some of the above functions, the functions other than those provided by the OS may be provided by a program.

[0149] The above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The flowcharts, sequences, elements included in the embodiments, and their arrangements, materials, conditions, shapes, sizes, etc., described in the embodiments are not limited to those illustrated and can be modified as appropriate. Furthermore, configurations shown in different embodiments can be partially substituted or combined with each other.

[0150] (Inventions Realized by the Present Embodiment) As described above, the present embodiment realizes the following inventions.

[0151] (Supplementary Note 1) An item list providing system comprising a server and a terminal, which generates a list in which a plurality of items are arranged in an order in which they are to be recommended to a user, presents the items included in the generated list to the user in the order in which they are arranged in the list, and when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifies, from the items in the list, items of low interest that are similar to the unnecessary item and have not yet been presented to the user, and the server and the terminal work together to perform a process of lowering the order of the low interest items in the list.

[0152] (Supplementary Note 2) A server comprising: a generating unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a providing unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a receiving unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed on the terminal to designate an unnecessary item from the items displayed on the screen; an identifying unit that identifies, from the items arranged in the list, low-interest items that are similar items to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an updating unit that lowers the order of the low-interest items in the list.

[0153] (Supplementary Note 3) The server according to Supplementary Note 2, wherein the report further includes display information indicating which items among the items included in the list are displayed on the screen of the terminal, and items not shown in the display information are regarded as hidden items.

[0154] (Supplementary Note 4) The server according to Supplementary Note 2 or 3, wherein an item that has not yet been provided to the terminal is set as the hidden item.

[0155] (Supplementary Note 5) The server according to any one of Supplementary Notes 2 to 4, which newly recommends a new item that is not similar to the unnecessary item and is not included in the list, and adds the new item to the list between the ranking of the items already provided to the terminal and the ranking of the items of low interest.

[0156] (Supplementary Note 6) A terminal comprising: an acquisition unit that acquires from a server a list in which multiple items are arranged in the order in which they are recommended; a display unit that displays the acquired list in a scrollable manner on a screen; a reception unit that receives an unnecessary selection operation that designates unnecessary items from the items arranged in the list and displayed on the screen; an identification unit that identifies low interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low interest items in the list.

[0157] (Supplementary Note 7) The terminal according to Supplementary Note 6, wherein the terminal reports the unnecessary item to the server, causing the server to lower a priority for recommending the reported item.

[0158] (Supplementary Note 8) The terminal according to Supplementary Note 6 or 7, wherein the terminal reports the unnecessary item to the server, thereby causing the server to identify similar items that are similar to the unnecessary item among the items in the list.

[0159] (Supplementary Note 9) The terminal according to any one of Supplementary Notes 6 to 8, further comprising: obtaining, from the server, a table of similarities between items in the list together with the list; and determining the similarities between each of the hidden items and the unnecessary items from the table.

[0160] (Supplementary Note 10) The terminal according to any one of Supplementary Notes 6 to 9, wherein a feature vector of each item in the list is acquired from the server along with the list, and a similarity between each of the hidden items and the unnecessary items is calculated based on the feature vector.

[0161] (Supplementary Note 11) The terminal according to Supplementary Note 9 or 10, wherein the similar items are identified by comparing the calculated similarity with a predetermined threshold, or a predetermined number or a predetermined percentage of items with the highest calculated similarity are identified as the similar items and the items of least interest.

[0162] (Supplementary Note 12) The terminal according to any one of Supplementary Notes 6 to 10, wherein the server newly recommends a new item that is not similar to the unnecessary item, and the new item is added to the list at a higher rank than the low-interest items.

[0163] (Supplementary Note 13) An item list providing method in an item list providing system having a server and a terminal, the item list providing method comprising: generating a list in which a plurality of items are arranged in an order in which they are to be recommended to a user; presenting items included in the generated list to the user in the order in which they are arranged in the list; when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifying low-interest items from the items in the list that are similar to the unnecessary item and have not yet been presented to the user; and performing a process in which the server and the terminal work together to lower the order of the low-interest items in the list.

[0164] (Appendix 14) A program that causes a computer to function as: a generating unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a providing unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a receiving unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed on the terminal to designate an unnecessary item from the items displayed on the screen; an identifying unit that identifies low-interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an updating unit that lowers the order of the low-interest items in the list.

[0165] (Appendix 15) A program that causes a computer to function as: an acquisition unit that acquires a list in which multiple items are arranged in the order in which they are recommended; a display unit that displays the acquired list in a scrollable manner on a screen; a reception unit that accepts an unnecessary selection operation to specify unnecessary items from the items arranged in the list and displayed on the screen; an identification unit that identifies low-interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the order of the low-interest items in the list.

[0166] According to the present invention, it is possible to provide an item list providing system, a server, a terminal, an item list providing method, and a program that can recommend items that are suitable for a user.

[0167] 10 Server 20 Terminal 30 Communication network 100 Item list providing system 11, 21 Processor 12, 22 Storage device 13, 23 Communication interface 14, 24 Input device 15, 25 Output device 110 Web browser 111 Item name 112 Detailed information of item 113 Recommendation list 114, 115 Scroll buttons RI1, RI2, RI3, RI4, RI5, RI6, RI8, RI9 Item (recommended item) DI Deleted item ds Unnecessary selection button rs Undo button

Claims

1. An item list providing system comprising a server and a terminal, which generates a list in which a number of items are arranged in an order in which they are recommended to a user, presents the items included in the generated list to the user in the order in which they are arranged on the list, and when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifies from the items in the list low-interest items that are similar to the unnecessary item and have not yet been presented to the user, and the server and the terminal work together to execute a process of lowering the order of the low-interest items in the list.

2. A server comprising: a generating unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a providing unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a receiving unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed to designate an unnecessary item from the items displayed on the screen of the terminal; an identifying unit that identifies low interest items among the items arranged in the list, which are similar items to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an updating unit that lowers the order of the low interest items in the list.

3. The server according to claim 2, wherein the report further includes display information indicating which of the items included in the list are displayed on the screen of the terminal, and items not shown in the display information are regarded as hidden items.

4. The server according to claim 2, wherein an item that has not yet been provided to the terminal is regarded as the hidden item.

5. A server as described in any one of claims 2 to 4, which newly recommends a new item that is not similar to the unnecessary item and is not included in the list, and adds the new item to the list between the ranking of the items already provided to the terminal and the ranking of the items of low interest.

6. A terminal comprising: an acquisition unit that acquires from a server a list in which multiple items are arranged in order of recommendation; a display unit that displays the acquired list in a scrollable manner on a screen; a reception unit that accepts an unnecessary selection operation to designate unnecessary items from among the items arranged in the list and displayed on the screen; an identification unit that identifies low interest items from among the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the ranking of the low interest items in the list.

7. The terminal according to claim 6, wherein the unnecessary item is reported to the server, causing the server to lower a priority for recommending the reported item.

8. The terminal according to claim 6, wherein the terminal reports the unwanted item to the server, thereby causing the server to identify items in the list that are similar to the unwanted item.

9. The terminal according to claim 6, further comprising: acquiring from said server, together with said list, a table of similarities between items in said list; and determining, from said table, said similarities between each of said hidden items and said unnecessary items.

10. The terminal according to claim 6, further comprising: acquiring from said server, together with said list, a feature vector of each item in said list; and determining a similarity between each of said hidden items and said unnecessary items based on said feature vector.

11. The terminal according to claim 9 or 10, wherein the similar items are identified by comparing the determined similarity with a predetermined threshold, or a predetermined number or a predetermined percentage of items having the highest determined similarity are identified as the similar items and the items of low interest.

12. The terminal according to any one of claims 6 to 10, further comprising: causing the server to newly recommend new items that are not similar to the unnecessary items; and adding the new items to the list in a higher order than the items of low interest.

13. An item list providing method in an item list providing system having a server and a terminal, comprising: generating a list in which a number of items are arranged in an order in which they are recommended to a user; presenting the items included in the generated list to the user in the order in which they are arranged in the list; when the user performs an unnecessary selection operation to designate an unnecessary item from the presented items, identifying low-interest items from the items in the list that are similar to the unnecessary item and have not yet been presented to the user; and performing a process of lowering the order of the low-interest items in the list in cooperation with the server and the terminal.

14. A program that causes a computer to function as: a generating unit that generates a list in which multiple items are arranged in the order in which they are recommended to a user; a providing unit that provides the items included in the generated list to a terminal according to the order in which they are arranged in the list and displays them on the screen of the terminal; a receiving unit that receives from the terminal a report indicating that an unnecessary selection operation has been performed to designate an unnecessary item from the items displayed on the screen of the terminal; an identifying unit that identifies low interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an updating unit that lowers the order of the low interest items in the list.

15. A program that causes a computer to function as: an acquisition unit that acquires a list in which multiple items are arranged in the order in which they are recommended; a display unit that displays the acquired list in a scrollable manner on a screen; a reception unit that accepts an unnecessary selection operation to specify unnecessary items from the items arranged in the list and displayed on the screen; an identification unit that identifies low interest items from the items arranged in the list that are similar to the unnecessary items and are hidden items that have not yet been displayed on the screen; and an update unit that lowers the ranking of the low interest items in the list.

Images