TWI812209B - Information processing system, information processing method and program product - Google Patents

Abstract

[Problem] We will propose products that are better for users. [Solution] The information processing system obtains a collection of multiple types that are related to each other. Based on the collection that has been obtained as mentioned above, the type that is related to the type of the item that the user has purchased or browsed is Select the related type; any one of the items belonging to the related type mentioned above that has been selected will be obtained as a recommended object.

Description

Information processing systems, information processing methods and program products

The present invention relates to an information processing system, an information processing method and a program product.

For example, in the sale of products through the Internet, there are systems that recommend products in response to user actions. In addition, as a recommendation target, products that are grouped with products that the user has purchased or browsed and have a high probability of being purchased will be selected.

Patent Document 1 discloses that products are recommended based on the user's action history based on their purchase history or browsing history. [Prior technical literature] [Patent Document]

[Patent Document 1] International Publication No. 2017/104064

[Problem to be solved by the invention]

A product that is frequently purchased when combined with a certain product may not necessarily be the best product for the user.

The present invention was developed in view of the above-mentioned problems, and its purpose is to provide a technology that proposes products that are more optimal for users. [Means used to solve problems]

In order to solve the above problems, the information processing system according to the present invention includes: a relationship acquisition means for acquiring a set of a plurality of types that are related to each other; and a type selection means for acquiring the set based on the above description. , and the type related to the type of the item that the user has purchased or browsed, that is, the related type, is selected; and the recommended object acquisition method is used to select the related type that belongs to the previously selected item. Any one of the items can be obtained as a recommended item.

In addition, the information processing method of the present invention includes the steps of: obtaining a collection containing a plurality of types that are related to each other; and based on the previously obtained collection, assigning items that the user has purchased or browsed to. The type of related type is the step of selecting the related type; and the step of obtaining any one of the items belonging to the previously selected related type as a recommendation object.

Furthermore, the program of the present invention causes the computer to function as: a correlation acquisition means for acquiring a set of a plurality of types that are related to each other; a type selection means for acquiring a set based on the above mentioned set, and Select the type associated with the type of the item that the user has purchased or browsed, that is, the related type; and the method for obtaining the recommended object is used to record the item belonging to the related type before it has been selected. Any one of them can be obtained as a recommendation object.

In one aspect of the present invention, the method for obtaining the recommendation object mentioned above is to use the item that has been purchased or viewed the most times among the items that belong to the related type mentioned above as the recommendation. object to obtain.

In one aspect of the present invention, the method for obtaining the recommended object mentioned above is to select an item that has been purchased or viewed the most times and is in stock among the items belonging to the related type mentioned above that have been selected. Get it as a recommendation object.

In one aspect of the present invention, the recommended target acquisition means mentioned above can acquire items that are on sale within a predetermined period as recommended targets.

In one aspect of the present invention, the means for obtaining the history mentioned above can obtain the purchase history or browsing history in a store different from the store the user is accessing as described above; the means for selecting the type mentioned above can be based on the set mentioned above. , and the type related to the type of items that the user mentioned above has purchased or browsed in other stores mentioned above is selected, which is the related type mentioned above; the method for obtaining recommended objects mentioned above can be used to obtain the items mentioned above Any item belonging to the previous store's item that is being accessed and has been selected as the previous item's related type will be acquired as the item recommended by the previous item.

In one aspect of the present invention, there is further provided a history acquisition means for acquiring a plurality of purchase histories each containing a combination of items that have been purchased together, or each containing a combination of items that have been browsed together. A plurality of browsing histories that form a combination; the means for obtaining the relationship mentioned above can be based on the type of the item that is composed of a plurality of combinations included in the plurality of purchase histories mentioned above or the plurality of browsing histories mentioned above. A collection of related types.

In one aspect of the present invention, the aforementioned history acquisition means acquires the aforementioned plurality of purchase histories in a plurality of stores or the aforementioned plurality of browsing histories in a plurality of stores; and the means for selecting the aforementioned type is based on the aforementioned plurality of purchase histories. The set has been determined, and the types related to the types of items that the user has purchased or browsed in the plurality of stores mentioned above, that is, the related types mentioned above, will be selected; the method for obtaining the recommended objects mentioned above is Any item among the plurality of stores mentioned above and any one of the items belonging to the related type mentioned above that has been selected as mentioned above can be obtained as the recommended target of the above mentioned items. [Effects of the invention]

Through the present invention, it is possible to propose products that are more optimal for users.

Hereinafter, embodiments of the present invention will be described based on the drawings. For components marked with the same symbols, repeated explanations will be omitted. This embodiment describes an information processing system that realizes e-commerce transactions at a plurality of stores. In this information processing system, products as items are proposed to users who have accessed a certain store.

FIG. 1 is a diagram illustrating an example of an information processing system according to an embodiment of the present invention. The information processing system includes an information processing server 1 and a client device 2 . The information processing server 1 is connected to one or a plurality of client devices 2 through a network.

The information processing server 1 includes a processor 11, a memory unit 12, a communication unit 13, and an input/output unit 14. In addition, the information processing server 1 is a server computer. The processing of the information processing server 1 can also be realized by a plurality of server computers. The client device 2 includes a processor 21, a memory unit 22, a communication unit 23, and an input/output unit 24. The client device 2 is a personal computer, a smart phone, or a tablet terminal.

The processors 11 and 21 operate according to the programs stored in the memory units 12 and 22. Furthermore, the processors 11 and 21 control the communication units 13 and 23 and the input/output units 14 and 24. In addition, the above-mentioned programs can be provided through the Internet, etc., or stored in computer-readable memory media such as flash memory or DVD-ROM.

The memory units 12 and 22 are composed of memory elements such as RAM and flash memory, and external memory devices such as hard disk drives. The memory units 12 and 22 store the above program. In addition, the memory units 12 and 22 store information or calculation results input from the processors 11 and 21, the communication units 13 and 23, and the input/output units 14 and 24.

The communication units 13 and 23 realize the function of communicating with other devices, and are composed of integrated circuits that implement wireless LAN or wired LAN, for example. The communication units 13 and 23 are based on the control of the processors 11 and 21 and input information received from other devices to the processors 11 and 21 or the memory units 12 and 22, and send the information to other devices.

The input/output units 14 and 24 are composed of a video controller that controls a display output device, a controller that obtains data from an input device, or the like. As input devices, there are keyboards, mice, touch panels, etc. The input/output units 14 and 24 output display data to the display output device based on the control of the processors 11 and 21, and obtain data input by the user by operating the input device. The display output device is, for example, an external display device.

Next, the functions provided by the information processing system are explained. Figure 2 is a block diagram of the functions implemented by the information processing system. The information processing system functionally includes: an overall history acquisition unit 51, a related type determination unit 52, a related item determination unit 53, a repurchase possibility acquisition unit 54, a user history acquisition unit 55, and a repurchase candidate acquisition unit 56 , related candidate acquisition unit 57 , list addition unit 58 , output unit 59 , and shopping cart control unit 60 . These functions are mainly realized by the processor 11 included in the information processing server 1 executing the program stored in the memory unit 12 and controlling the communication unit 13 and so on. Even part of the function of the output unit 59 can be realized by the processor 21 included in the client device 2 executing the program stored in the memory unit 22 to control the communication unit 23 and the input/output unit 24. The related candidate acquisition unit 57 functionally includes a type utilization unit 61 and an item utilization unit 62 . Functionally, the type utilization unit 61 includes a type selection unit 66 and an item selection unit 67 , and the item utilization unit 62 includes an item selection unit 69 .

The overall history acquisition unit 51 acquires at least one of a purchase history and a browsing history. Each purchase history includes a combination of information indicating a user who has made a purchase and items that have been purchased together. Each browsing history includes a combination of information indicating users who have browsed, and items that have been browsed together. For example, items that have been viewed within a predetermined period (such as on the same day) can be regarded as items that have been viewed together. The purchase history and browsing history are stored in the memory unit 12 for a plurality of users who use the information processing system, for example, all users. The overall history acquisition unit 51 may acquire all of the purchase history and browsing history stored in the memory unit 12, or may acquire the purchase history and browsing history of a specific store.

The related type determination unit 52 determines a set of plural types that are related to each other based on the type to which the items constituted by the plural combinations included in the plural purchase histories or the plural browsing histories that have been acquired.

The related item determination unit 53 determines a set of a plurality of items related to each other based on items composed of a plurality of combinations included in a plurality of acquired purchase histories or a plurality of browsing histories.

The repurchase possibility acquisition unit 54 calculates the repurchase possibility for each item based on the acquired purchase history.

The user history acquisition unit 55 acquires an action history including at least one of a purchase history including items purchased by the user and a browsing history including items viewed by the user.

The repurchase candidate acquisition unit 56 selects at least one of the items included in the purchase history of the user as a repurchase candidate.

The related candidate acquisition unit 57 selects at least one item that is different from the items included in the purchase history and browsing history of the user as a related candidate. The method in which the repurchase candidate acquisition unit 56 acquires items as repurchase candidates is different from the method in which the related candidate acquisition unit 57 selects items as related candidates.

The category utilization unit 61 included in the related candidate acquisition unit 57 selects categories related to the categories of items that the user has purchased or browsed, and selects the items belonging to the selected categories. Any one of them is obtained as a recommendation target, that is, a related candidate.

The type selection unit 66 included in the type utilization unit 61 selects a type related to the type of the item that the user has purchased or browsed based on the acquired collection, that is, a related type.

The item selection unit 67 included in the category utilization unit 61 acquires any item included in the selected category as a recommendation target, that is, a related candidate. More specifically, the item selection unit 67 acquires the item that has been purchased or viewed the most times among the items belonging to the selected category as a related candidate. The item selection unit 67 may also acquire items that have started to be sold within a predetermined period as related candidates. Here, the predetermined period may be a period in the past one month, one week, etc.

The item utilization unit 62 included in the related candidate acquisition unit 57 selects items related to items that the user has purchased or browsed, and uses the selected items as recommendation targets. Related candidates are obtained.

The item selection unit 69 included in the item utilization unit 62 acquires items related to items that the user has purchased or browsed based on the acquired set as related candidates.

The list adding unit 58 adds the selected repurchase candidates and related candidates to the list. The list adding unit 58 adds the repurchase candidates and the related candidates to the list in such a manner that the repurchase candidates are placed earlier than the related candidates.

The output unit 59 displays an image in which the items included in the list are arranged at positions corresponding to their order. When the output unit 59 is configured only with the information processing server 1 , the image data is sent to the client device 2 to be displayed. In addition, when the output unit 59 is also included in the client device 2, the image is displayed on the display device of the client device 2.

The shopping cart control unit 60 adds items that are included in the list and have been indicated by the user to the shopping cart.

Next, the details of the processing of the information processing system will be described. FIG. 3 is a flowchart showing an example of the processing of the overall history acquisition unit 51 , the related type determination unit 52 , the related item determination unit 53 , and the repurchase possibility acquisition unit 54 . The processing shown in FIG. 3 only needs to be performed once in advance, but may be repeatedly performed at predetermined intervals, for example.

First, the overall history acquisition unit 51 acquires purchase histories and browsing histories of a plurality of users (step S101). The overall history acquisition unit 51 can acquire the purchase history and browsing history stored in the memory unit 12 for a certain store regardless of the user, or it can acquire all the stores managed by the information processing system. The purchase history and browsing history stored in the memory unit 12 are obtained. In addition, the purchased purchase history and browsing history obtained may also be the purchase history and browsing history in the past year.

Next, the correlation type determination unit 52 extracts a combination of items that have been purchased together from the acquired purchase history, and extracts a combination of items that have been browsed together from the acquired browsing history. combination (step S102). Then, the related type determination unit 52 obtains the type to which each item included in the extracted combination belongs (step S103). Strictly speaking, the type system may have a hierarchical structure, and the related type determination unit 52 may obtain the type at the lowest level (in the case of a three-level type, the type at the third level). The type used here may be a type in a certain store, or may be a common type among all stores.

Then, the related type determining unit 52 determines a set of mutually related types based on the type to which each item included in the combination belongs (step S104). The related type determination unit 52 counts the types of sets to which the items included in the combination belong, and determines the sets of mutually related types based on the number of the counted types of sets. For example, the correlation type determination unit 52 may determine the set of types to be a set of mutually related types when the number of counted types is greater than a threshold.

Furthermore, the related item determination unit 53 determines a set of mutually related items based on the items included in the extracted combination (step S105). For example, the related item determination unit 53 may count the number of occurrences of a combination and determine a set of mutually related items based on the counted number of times. For example, when the number of occurrences of a counted combination is greater than a threshold, the related item determination unit 53 may determine the items included in the combination into a set of mutually related items.

In addition, the related type determining unit 52 may also determine a set of mutually related items that has been determined by the related item determining unit 53, and determine the set of the types to which each of the items included in the set belongs. , determined to become a collection of interrelated types.

Furthermore, the repurchase possibility acquisition unit 54 calculates the repurchase possibility for each item (step S106). For example, the repurchase possibility may be the ratio of users who purchased the item to be calculated multiple times among users who purchased the item to be calculated, which may be regarded as the repurchase rate. In addition, the repurchase possibility acquisition unit 54 may store items whose repurchase possibility is greater than a predetermined condition, that is, a threshold, in the memory unit 12 as items with a high repurchase possibility. middle.

In addition, the repurchase possibility can also be predicted using a learning model. The repurchase possibility acquisition unit 54 acquires the repurchase possibility based on the past purchase date of the item by the user. Specifically, the repurchase possibility acquisition unit 54 uses a learning model that takes the past purchase date of the item as input and then outputs the repurchase possibility of the item to obtain the repurchase possibility of the item. The learning model utilizes well-known machine learning algorithms such as neural networks, and uses the relationship between the past purchase date of the item purchased by the user and the presence or absence of repurchase by the user as teacher data. Supervised machine learning to learn and be generated. The repurchase possibility acquisition unit 54 may also regard the items whose repurchase possibility satisfies the predetermined conditions among the items included in the user's purchase history as the items with high repurchase possibility. and stored in the memory unit 12 . For example, if the learning model outputs the repurchase possibility in the form of a numerical value, the item whose repurchase possibility is greater than the predetermined condition, that is, the threshold, can also be regarded as an item with a high repurchase possibility. . In addition, when the learning model outputs yes or no as the repurchase possibility, the item whose repurchase possibility meets a predetermined condition (for example, yes) can also be regarded as an item with a high repurchase possibility. .

There is a certain relationship between the item's past purchase date and the repurchase possibility. For example, items that the user purchased recently or items that the user purchased long ago have a low repurchase possibility because the user's purchase desire at the current point is low. On the other hand, for items that users purchase regularly, the probability of repurchase of the items in the next purchase period is higher. By using a learning model that learns the past purchase date of the item and whether it has been repurchased, the user's repurchase possibility can be determined with high accuracy.

In addition, the input data of the learning model is not limited to the past purchase date of the item. For example, the input data may also be the repurchase rate. This enables predictions that reflect the actual repurchase results for each item. For example, the input data may also be the number of purchases of items in the past caused by the user. In this way, predictions can be made that reflect the actual purchase results of each user's items. For example, the input data may also be the number of items purchased last time by the user. This makes it possible to predict the period between the previous purchase period and the next purchase period, reflecting changes in the number of items purchased last time. For example, the input data can also be the price of the item. This makes it possible to predict the price of items that reflect the low likelihood of repurchasing high-priced items. For example, the input data may also be the repurchase rate of the type to which the item belongs. This makes it possible to predict the repurchase rate of the type that is correlated with the repurchase rate of the item, even for an item whose actual purchase results are small. In addition, multiple input data among the above-mentioned input data can also be used as input data for the learning model. By using a learning model that has learned the relationship between the above input data and the user's repurchase as teacher data, the user's repurchase possibility can be determined with high accuracy.

Next, the processing when the user accesses the sales page of a certain store in the information processing system will be described. Figures 4 to 6 are diagrams illustrating an example of processing related to sales to users. 4 to 6 illustrate an outline of the processing of the user history acquisition unit 55, the repurchase candidate acquisition unit 56, the related candidate acquisition unit 57, the list addition unit 58, the output unit 59, and the shopping cart control unit 60.

First, the user history acquisition unit 55 identifies the user who has accessed the information processing server 1 (step S201), and acquires the purchase history and browsing history of the store accessed by the user. (Step S202).

When the purchase history has been acquired (Y in step S203), the repurchase candidate acquisition unit 56 selects the item with the highest repurchase possibility among the items included in the acquired purchase history. The item is selected as a repurchase candidate, and the list adding unit 58 adds the selected item to the list (step S204). In addition, the type utilization unit 61 included in the related candidate acquisition unit 57 selects candidate items from a plurality of items displayed in the store based on the purchase history and the related type, and the list adding unit 58 will have been selected. The candidate items are added to the list (step S205). The item utilization unit 62 included in the related candidate acquisition unit 57 selects candidate items from a plurality of items displayed in the store based on the purchase history and the related items, and the list adding unit 58 will have been selected. The candidate items are added to the list (step S206). Here, the list adding unit 58 adds the items so that the order of the items added in step S204 is earlier than the order of the items added in steps S205 and S206. Append. Once the processing of steps S204 to S206 is performed, the processing proceeds to step S219.

The processing of step S205 will be described in detail. FIG. 7 is a diagram illustrating an example of processing by the type utilization unit 61. First, the type selecting unit 66 included in the type using unit 61 acquires the type of each item included in the action history consisting of the purchase history or browsing history (step S301). For example, in step S205, the action history is a purchase history. Then, the type selection unit 66 selects a type (related type) related to the acquired type based on the set of plural types related to each other determined by the related type determining unit 52 (step S302 ).

The item selection unit 67 selects the highest-ranked item as a candidate item among the items belonging to the related type and displayed in the store (step S303). The order can be set based on its sales quantity or sales amount in that type. Furthermore, the item selection unit 67 deletes items that overlap with items selected from other action histories from the selected items (step S304). Then, the item selecting unit 67 selects items that are still in stock among the selected items as candidate items (step S305), and the list adding unit 58 adds the candidate items to the list. .

The processing of step S206 will also be described. FIG. 8 is a diagram illustrating an example of processing by the item utilization unit 62 . First, the item selection unit 69 included in the item utilization unit 62 selects each item included in the action history based on the set of a plurality of mutually related items that have been determined by the related item determination unit 53. The items are related and the items displayed in the store are selected (step S401). The item selection unit 69 deletes duplicate items from the selected items (step S402), and selects items that are still in stock among the selected items as candidate items (step S402). Step S403). The list adding unit 58 adds candidate items to the list.

Here, if the purchase history is not acquired in step S202 (N in step S203), the process proceeds to step S207.

When the browsing history has been acquired (Y in step S207), the type utilization unit 61 included in the related candidate acquisition unit 57 selects the item as a candidate item based on the browsing history and the related type. , the list adding unit 58 adds the selected candidate items to the list (step S208). The details of step S208 are based on changing the action history into browsing history in the process shown in FIG. 7 . The item utilization unit 62 included in the related candidate acquisition unit 57 selects items as candidate items based on the browsing history and related items, and the list addition unit 58 adds the selected candidate items. to the list (step S209). The details of the process of step S209 are to change the action history into the browsing history in the process shown in FIG. 8 . Once the processes of steps S208 and S209 are performed, the process proceeds to step S219.

Here, if the browsing history is not obtained in step S202 (N in step S207), the process proceeds to step S210.

The processes described in steps S210 to S217 are processes performed when the user does not have a purchase history or a browsing history in the store accessed by the user.

In step S210, the user history acquisition unit 55 acquires the user's purchase history and browsing history in other stores (step S210). Here, when neither the purchase history nor the browsing history can be obtained (N in step S211), since recommendations based on the purchase history and browsing history cannot be made, the list adding unit 58 adds stores accessed by the user. Items included in the sales ranking list are added to the list (step S212).

On the other hand, when the purchase history has been acquired (Y in step S211 and Y in step S213), the type utilization unit 61 included in the related candidate acquisition unit 57 is based on the purchase history and related types in other stores. , select candidate items from the items displayed in the store, and add the selected candidate items to the list (step S215). The process of step S215 is to change the action history to the purchase history in other stores in the process shown in FIG. 7 . If the purchase history has not been acquired (Y in step S211 and N in step S213), step S215 is skipped.

In addition, when the browsing history has been acquired (Y in step S211 and Y in step S216), the type utilization unit 61 included in the related candidate acquisition unit 57 is based on the browsing history and related types in other stores. Select candidate items from the items displayed in the store, and add the selected candidate items to the list (step S217). The process of step S217 is to change the action history to the browsing history in other stores in the process shown in FIG. 7 . When the browsing history has not been acquired (Y in step S211 and N in step S216), step S217 is skipped and the process proceeds to step S219.

Here, a specific example will be described regarding the items added to the list in steps S202 to S217.

FIG. 9 is a diagram illustrating an example of the relationship between the purchase history in store S and items added to the list. Store S is assumed to be a store accessed by the user. In the case shown in Figure 9, the user's purchase history exists in store S, and the items included in the purchase history are kitchen towel A, toilet paper B, diaper C, and mineral water D. The list includes items acquired through the repurchase candidate acquisition unit 56 (corresponding to the first selection method), namely, kitchen towel A, toilet paper B, and disposable diaper C (refer to "Repurchase Proposal" in Fig. 9) , and the items selected by the type utilization unit 61 included in the related candidate acquisition unit 57 (corresponding to the second selection method) are milk powder E, baby food F, and thermos G (refer to "Related Type" in Fig. 9 Proposal"), and the items selected by the item utilization unit 62 included in the related candidate acquisition unit 57 (corresponding to the second selection method) are mineral water H and mineral water J 12-bottle collection pack (Refer to "Proposals Caused by Related Items"). In Figure 9, it is assumed that the items located at the top are ranked higher.

Kitchen paper towel A, toilet paper B, and disposable diaper C that have been added to the list are items included in the user's purchase history in store S that are more likely to be repurchased. Mineral water D, which is less likely to be repurchased, was not added to the list.

Diaper C belongs to the type of "disposable diapers", milk powder E, baby food F, and thermos bottle G belong to the types of "milk powder", "baby food" and "baby products" respectively. The type of "disposable diaper" is related to the types of "milk powder", "baby food", and "baby supplies", and the type utilization unit 61 of the related candidate acquisition unit 57 is related to the type of "disposable diaper". Among them, the items with the highest ranking, that is, milk powder E, baby food F, and thermos bottle G, are selected as candidates for proposal. Here, when the item with the highest ranking is not in stock, the type utilization unit 61 does not select the item as a candidate. In this case, the type utilization unit 61 may select the highest-ranking item among the items still in stock as a candidate. In addition, the item utilization unit 62 of the related candidate acquisition unit 57 may select the mineral water H and the mineral water J 12-bottle collection package related to the mineral water D as candidates for proposal. Here, the item utilization unit 62 does not select items that are not in stock as candidates.

By selecting items to propose based not just on related items but also on related types, the range of items proposed to the user becomes larger and the items proposed are more relevant. The possibility will become higher depending on the user's condition. In this way, the possibility of the item being purchased can be increased, while the user's trouble of retrieving the item can be reduced, and the ease of use can be improved. In addition, the information processing server that has been input with items that the user has purchased or browsed automatically outputs items to be proposed by referring to a collection of multiple types that are related to each other. This makes it possible to obtain items to be proposed to the user at high speed and with high accuracy. In addition, compared with the number of items, the number of types is small, so the amount of data processed by the information processing server can be reduced, and the processing load can be reduced.

FIG. 10 is a diagram illustrating an example of the relationship between the purchase history and browsing history in store S and items added to the list. In the case shown in Figure 10, there is no purchase history of the user in store S, but there is a browsing history. The items included in the browsing history are coffee K and makeup remover L. The type "makeup remover" to which makeup remover L belongs is related to the type "sunscreen oil" to which sunscreen oil O belongs, so the items that will become candidates are selected like this. In this case as well, the selection of proposed items based on the mutually related types of the type utilization unit 61 and the selection of the proposed items based on the mutually related items of the item utilization unit 62 are performed. The selection of candidates is performed by two methods (which may also correspond to the first selection method and the second selection method). In addition, even if the user's purchase history exists in store S, candidates can still be selected based on the browsing history.

FIG. 11 is a diagram illustrating an example of the relationship between purchase history and browsing history and items added to the list. In the case shown in Figure 11, the user's purchase history and browsing history do not exist in store S, but there are purchase histories and browsing histories in other stores. The purchase history from other stores contains diaper C, and the browsing history contains makeup remover L. In this case, the type utilization unit 61 of the related candidate acquisition unit 57 selects candidates by two selection methods: selection of candidates based on purchase history and selection of candidates based on browsing history. In addition, even if the user's purchase history or browsing history exists in store S, candidates can still be selected based on the purchase history or browsing history in other stores.

In addition, the information processing server 1 can select candidate items in store S through various selection methods based on purchase history or browsing history in other stores. For example, if store S displays items included in the purchase history or browsing history of users in other stores, it can also display items included in the purchase history or browsing history of users in other stores. Items with the same item are selected as candidate items in store S. At this time, among the items included in the purchase histories of users in other stores, items with a high repurchase possibility may be selected as candidate items in store S. For example, the type utilization unit 61 included in the related candidate acquisition unit 57 may also combine the candidate items in the store S with mutually related types based on the purchase history or browsing history of users in other stores. select. For example, the item utilization unit 62 included in the related candidate acquisition unit 57 may also use items that are related to each other based on the purchase history or browsing history of users in other stores. items to select.

In the store that the user is accessing, candidate items are selected based on the purchase history or browsing history of users in other stores. This way, the purchase history or browsing history of users in other stores can be considered. Propose appropriate items. In this way, the possibility of the item being purchased can be increased, while the user's trouble of retrieving the item can be reduced, and the ease of use can be improved.

In step S219, the output unit 59 outputs the proposal screen including the list to the display device owned by the client device 2 or connected to the client device 2. Here, the output unit 59 outputs the proposal screen so that the candidate items included in the list are arranged at positions corresponding to their order. For example, the proposal screen is a screen displayed on the web page of the store accessed by the user.

FIG. 12 is an illustration of an example of an output proposal screen. A plurality of add buttons 81 and a checkout button 84 are arranged in the proposal screen, and a plurality of items included in the list are arranged. Each item is configured with its own add button 82 and check box 83 .

The individual add button 82 is a button used to allow the user to instruct the user to add the corresponding item to the shopping cart. The check box 83 is a button required to check the corresponding item, and the plural add button 81 is used to instruct the item corresponding to the check box 83 to be checked when pressed. Add to cart button required. The checkout button 84 is a button necessary to instruct the start of the purchase process for the item that has been added to the shopping cart. Once either the individual add button 82 or the checkout button 84 is pressed, the client device 2 sends information indicating this fact to the information processing server 1 . Once the plurality of add buttons 81 are pressed, the client device 2 will provide information indicating the item corresponding to the check box 83 that was checked when it was pressed, and information indicating the fact that it was pressed. , and sent together to information processing server 1.

On the proposal screen, candidate items selected by a plurality of methods and added to the same list, such as repurchase candidates and candidates based on relationship types, are integrated and displayed. In this way, the places that users refer to can be concentrated in one place, which can reduce the load of reference products. In addition, the information processing server sends the candidate items selected through multiple methods to the client device not as data in different lists but as data in the same list. This can reduce the amount of data sent by the information processing server to the client device, thereby reducing the communication load and speeding up the display processing of the proposal screen in the client device.

Steps S221 to S229 are processes performed when the user operates the proposal screen. This processing is explained below.

In step S221, the shopping cart control unit 60 determines whether the individual add button 82 has been pressed by the user. When the individual add button 82 has been pressed (Y in step S221), the shopping cart control unit 60 adds the item corresponding to the pressed individual add button 82 to the shopping cart (step S222). . Then, the process proceeds to step S225, and the output unit 59 changes the individual add button 82 of the item added to the shopping cart into the quantity input field 85 (see FIG. 13). The change to the quantity input field 85 can be realized by the client device 2 executing a program that is sent together with the data on the proposal screen, and replacing the currently displayed individual add button 82 with the quantity input field 85. This can also be realized by the information processing server 1 sending the data of the proposal screen in which the individual add button 82 has been replaced with the quantity input field 85 to the client device 2, and the client device 2 displaying the proposal screen.

FIG. 13 is an illustration of an example of a proposal screen after items are added to the shopping cart. In the example of FIG. 13 , when the individual add button 82 is pressed for the upper left item among the proposed items in FIG. 12 , the individual add button 82 changes to The quantity input field looks like 85.

When the individual add button 82 has not been pressed (N in step S221), the shopping cart control unit 60 determines whether the plural add buttons 81 have been pressed (step S223). When the plurality of add buttons 81 have been pressed (Y in step S223), the shopping cart control unit 60 adds the items whose corresponding check boxes 83 have been checked to the shopping cart (step S224). Then, in step S225, the output unit 59 changes the individual add button 82 of the item added to the shopping cart into the quantity input field 85.

Once the processing of step S225 is performed, the processing after step S221 will be repeated.

On the other hand, when the plural add button 81 is not pressed (N in step S223), the shopping cart control unit 60 determines whether the quantity input field 85 has been changed (step S226). When the quantity input field 85 has been changed (Y in step S226), the shopping cart control unit 60 changes the quantity in the shopping cart of the item corresponding to the changed quantity input field 85 (step S227). ). Thereafter, the processing from step S221 onward is repeated.

If the quantity input field 85 has not been changed (N in step S226), the shopping cart control unit 60 determines whether the checkout button 84 has been pressed (step S228). When the checkout button 84 has been pressed (Y in step S228), the shopping cart control unit 60 performs processing related to the purchase of items existing in the shopping cart, more specifically, performs checkout. and delivery-related processing (step S229). On the other hand, when the checkout button 84 is not pressed (N in step S228), the processing from step S221 onward is repeated.

Although the embodiment of the present invention has been described, the application of the present invention is not necessarily limited to this embodiment. For example, items are not only physical products that are sold, but may also be content that is sent as data.

1:Information processing server 2: Client device 11: Processor 12:Memory Department 13:Communication Department 14:Input and output department 21: Processor 22:Memory Department 23: Ministry of Communications 24:Input/output department 51: Overall resume acquisition department 52:Relationship Type Determination Department 53: Related Item Decision Department 54:Repurchase possibility acquisition department 55: User history acquisition department 56: Repurchase candidate acquisition part 57: Related Candidate Acquisition Department 58: List addition department 59:Output Department 60: Shopping Cart Control Department 61: Type Utilization Department 62:Item Utilization Department 66:Type selection department 67:Item Selection Department 69:Item Selection Department 81:Plural add button 82:Individual add button 83: Check box 84: Checkout button 85:Quantity input field

[Fig. 1] A diagram illustrating the hardware configuration of the information processing system according to the embodiment of the present invention. [Figure 2] Block diagram of the functions implemented by the information processing system. [Fig. 3] A flowchart of an example of processing by the overall history acquisition unit, the related type determination unit, the related item determination unit, and the repurchase possibility acquisition unit. [Fig. 4] An illustration of an example of processing related to sales to users. [Fig. 5] An illustration of an example of processing related to sales to users. [Fig. 6] An illustration of an example of processing related to sales to users. [Fig. 7] A diagram illustrating an example of processing by the type utilization unit. [Fig. 8] A diagram illustrating an example of processing by the item utilization unit. [Fig. 9] A diagram illustrating an example of the relationship between the purchase history in store S and items added to the list. [Fig. 10] An illustration of an example of the relationship between the purchase history and browsing history in store S and items added to the list. [Fig. 11] An illustration of an example of the relationship between purchase history and browsing history and items added to the list. [Fig. 12] An illustration of an example of an output proposal screen. [Figure 13] An illustration of an example of a proposal screen after items are added to the shopping cart.

Claims (10)

An information processing system includes: a relationship acquisition unit, which is used to obtain a set of multiple types that each of a plurality of items that have been purchased or browsed together belongs to, and is a set of multiple types that are related to each other; and type selection. The part is used to select the type related to the type of the item that the user has purchased or browsed based on the previously obtained collection, that is, the related type; and the recommendation object acquisition part is used to select Any of the items belonging to the previously selected related type will be obtained as a recommendation object; the type selection part of the preceding paragraph is based on the set of the preceding paragraph, and will be different from the store that the user of the preceding paragraph is accessing. In other stores, the type of items that the user has purchased or browsed belongs to, which is the related type mentioned above, and is selected; the recommended object acquisition part mentioned above is to obtain the store mentioned above where the user is currently accessing. Any of the items belonging to the related type of the previous item that has been selected as the previous item will be obtained as the recommended target of the previous item. An information processing system as described in claim 1, wherein the recommended object acquisition unit mentioned above is an item that has been purchased or viewed the most times among the items belonging to the related type mentioned above that have been selected. , obtained as a recommendation object. For example, the information processing system described in claim 1, which , the recommendation object acquisition part mentioned above is to acquire the items that have been purchased or viewed the most times and are in stock among the items that belong to the selected related type mentioned above as the recommended objects. . The information processing system described in any one of Claims 1 to 3, wherein the recommendation target acquisition unit mentioned above acquires items that are on sale within a predetermined period as recommendation targets. An information processing system includes: a relationship acquisition unit, which is used to obtain a set of multiple types that each of a plurality of items that have been purchased or browsed together belongs to, and is a set of multiple types that are related to each other; and type selection. The part is used to select the type related to the type of the item that the user has purchased or browsed based on the previously obtained collection, that is, the related type; and the recommendation object acquisition part is used to select Any one of the items belonging to the related type of the previous item that has been selected is acquired as a recommendation object; and the resume acquisition part is used to acquire items that respectively include items that have been purchased together. A plurality of purchase histories that are combined, or a plurality of browsing histories that each include a combination of items that have been browsed together; the relationship acquisition part mentioned above is based on combining the plurality of purchase histories mentioned above or the plurality of browsing histories mentioned above. The type of items composed of a plurality of combinations determines the set of the plurality of types that are related to each other. combine. An information processing system as described in claim 5, wherein the aforementioned history acquisition unit acquires the aforementioned plurality of purchase histories in a plurality of stores, or the aforementioned plurality of browsing histories in a plurality of stores; the aforementioned history type selection Based on the determined set mentioned above, the type related to the types of items that the user has purchased or browsed in the plurality of stores mentioned above, that is, the related type mentioned above, is selected; recommended by the above mentioned The object acquisition unit obtains any item among the plurality of stores mentioned above and any one of the items belonging to the previously mentioned related type that has been selected as the aforementioned recommendation object. An information processing method, which includes the steps of: obtaining, by a computer, a set of a plurality of types that each of a plurality of items that have been purchased or browsed together belongs to, and is a collection of mutually related types; and by the computer based on the foregoing description The collection has been obtained, and the type associated with the type of the item that the user has purchased or browsed is the step of selecting the related type; and the step of selecting the previously selected related type by the computer The step of obtaining any of the items that belong to it as a recommendation object; in the step of selecting the relationship type mentioned above, it is based on the set mentioned above, and other stores that are different from the store the user is accessing are included in the step mentioned above. should be used The type related to the type of items that the user has purchased or browsed is the aforementioned related type, and is selected; in the step of obtaining the aforementioned recommended objects, the items in the aforementioned store where the user is currently accessing are selected. The item is any one of the items belonging to the related type of the previous item that has been selected, and is obtained as the recommended target of the previous item. An information processing method, which includes the steps of: obtaining, by a computer, a set of a plurality of types that each of a plurality of items that have been purchased or browsed together belongs to, and is a collection of mutually related types; and by the computer based on the foregoing description The collection has been obtained, and the type associated with the type of the item that the user has purchased or browsed is the step of selecting the related type; and the step of selecting the previously selected related type by the computer The step of obtaining any one of the items that belong to it as a recommendation object; and obtaining from the computer a plurality of purchase histories each containing a combination of items that have been purchased together, or each containing a combination of items that have been purchased together. The step of selecting a plurality of browsing histories that are combinations of items that are browsed together; the step of selecting the aforementioned relationship type is based on a plurality of combinations included in the aforementioned plurality of purchase histories or the aforementioned plurality of browsing histories. The type of the item belongs to determines the set of multiple types that are related to each other. A computer program product, which is characterized by being used to make a computer function: The related acquisition part is used to obtain the type of each of the plurality of items that have been purchased or browsed together, and is a set of multiple types that are related to each other; the type selection part is used to obtain the set of multiple types that have been purchased or browsed together based on the above. The obtained set is used to select the type related to the type of the item that the user has purchased or browsed, that is, the related type; and the recommended object acquisition part is used to record the related type before the previous item has been selected. Any one of the items belonging to the above mentioned items is obtained as a recommendation object; the type selection part mentioned above is based on the above mentioned collection, and selects the items that the user should have visited in other stores that are different from the store the user is accessing. The type associated with the type of the item purchased or browsed is the related type mentioned above, and is selected; the recommended object acquisition part mentioned above is to select the item in the previous store where the user is accessing and the item mentioned above has been selected. Any of the items belonging to the related types mentioned above will be acquired as the recommended objects mentioned above. A computer program product, which is characterized by being used to make a computer function: a related acquisition unit is used to obtain the type of each of a plurality of items that have been purchased or browsed together, and are related to each other. A collection of multiple types; the type selection unit is used to select, based on the previously obtained collection, types related to the types of items that the user has purchased or browsed, that is, related types; and The recommended object acquisition part is used to acquire any of the items belonging to the previously selected related types as recommended objects; the resume acquisition part is used to acquire the items that have been selected together. A plurality of purchase histories that are combinations of purchased items, or a plurality of browsing histories that each include a combination of items that have been viewed together; the related acquisition part mentioned above is based on combining the plurality of purchase histories or items mentioned above. The type of items composed of the plurality of combinations contained in the plurality of browsing histories determines the set of the plurality of types that are related to each other.

Images