Information processing apparatus, information processing method, and information processing program

Abstract

To provide information having a high degree of harmony.SOLUTION: An information processing apparatus includes a calculation unit and a provision unit. The calculation unit calculates a set matching score indicating a degree of harmony between a target group, user data, and additional information, on the basis of the target group formed by combining a plurality of targets, the user data, and the additional information. The provision unit provides information on the combination of the target group, the user data, and the additional information, the combination satisfying a predetermined condition, on the basis of the set matching score calculated by the calculation unit.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] A technique called set matching is known that calculates the matching score between sub-coordinates. The higher the matching score, the better the compatibility between the sub-coordinates. This makes it possible to select the optimal combination from a large number of options when proposing outfits. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-153195 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the conventional technology, since set matching is performed on two axes between sub-codes, there is room for further improvement in order to provide information with a high degree of harmony.

[0005] The present application has been made in view of the above, and aims to provide information with a high degree of harmony. [Means for solving the problem]

[0006] The information processing device of the present application is characterized by having a calculation unit that calculates a set matching score indicating the degree of harmony between the object group, the user data, and the additional information based on an object group combining multiple objects, user data, and additional information, and a provision unit that provides information regarding a combination of the object group, the user data, and the additional information that satisfies a predetermined condition based on the set matching score calculated by the calculation unit. Effect of the Invention

[0007] According to one aspect of the embodiment, it is possible to provide information with a high degree of harmony. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an information processing system according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of information processing according to the embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of the configuration of the information display device according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Diagram 5] FIG. 5 is a diagram illustrating an example of a posted information storage unit according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a model information storage unit according to the embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of a user information storage unit according to the embodiment. [Figure 8] FIG. 8 is a flowchart illustrating an example of information processing according to the embodiment. [Figure 9] FIG. 9 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the information processing device, the information processing method, and the information processing program according to the present application will be described in detail with reference to the drawings. Note that the information processing device, the information processing method, and the information processing program according to the present application are not limited to the embodiments. In addition, the same parts in the following embodiments are given the same reference numerals, and duplicated explanations are omitted.

[0010] (Embodiment) [1. Configuration of information processing system] An information processing system 1 shown in FIG. 1 will be described. As shown in FIG. 1, the information processing system 1 includes an information display device 10 and an information processing device 100. The information display device 10 and the information processing device 100 are connected to each other via a predetermined communication network (network N) so as to be able to communicate with each other by wire or wirelessly. FIG. 1 is a diagram showing an example of the configuration of the information processing system 1 according to an embodiment. Note that the information processing system 1 shown in FIG. 1 may include a plurality of information display devices 10 and a plurality of information processing devices 100.

[0011] The information display device 10 is an information processing device used by a user. The information display device 10 may be any device that can realize the processing in the embodiment. The information display device 10 may also be a device such as a smartphone, a tablet terminal, a notebook PC, a desktop PC, a mobile phone, or a PDA. In the example shown in FIG. 2, the information display device 10 is a smartphone.

[0012] The information display device 10 is, for example, a smart device such as a smartphone or a tablet, and is a mobile terminal device capable of communicating with any server device via a wireless communication network such as 3 to 5G (Generation) or LTE (Long Term Evolution). The information display device 10 has a screen such as a liquid crystal display having a touch panel function, and may receive various operations on display data such as content, such as a tap operation, a slide operation, a scroll operation, etc., from a user using a finger or a stylus.

[0013] The information processing device 100 is an information processing device that aims to provide information with a high degree of harmony. For example, the information processing device 100 is an information processing device for proposing the selection of an optimal combination through set matching. The information processing device 100 also has a function of calculating a matching score (corresponding to a set matching score) by set matching of three axes in which one axis of coordination is added with user data and additional information. The information processing device 100 is also realized by, for example, a server device or a cloud system. The information processing device 100 is also an information processing device such as a PC or a WS (Work Station), and performs processing based on information transmitted from the information display device 10 or the like via a network N.

[0014] The information processing device 100 also provides an electronic commerce service (mail order site, electronic shopping mall, auction site, flea market site, etc.) for searching, viewing, purchasing, etc. of items (also called products) including fashion items (tops, jackets, pants, skirts, bags, shoes, fashion accessories, wallets, watches, accessories, underwear, legwear, caps, hats, cosmetics, etc.), interior items (cushions, room shoes, rugs, candles, furniture, lighting, table clocks, wall clocks, interior accessories, photo frames, flower vases, posters, art, etc.), digital items (digital fashion items, digital interior items, etc.). The information processing device 100 also provides a coordination service (coordination site, etc.) for posting, searching, viewing, virtually trying on, etc., images (still images, videos, etc.) showing fashion coordination and images (still images, videos, etc.) showing makeup. The information processing device 100 may also provide a matching service (matching site, etc.) for searching, viewing, etc., for an ideal user. The information processing device 100 may also provide a body management service for managing body shapes, etc.

[0015] Although FIG. 1 shows a case where the information display device 10 and the information processing device 100 are separate devices, the information display device 10 and the information processing device 100 may be integrated.

[0016] [2. An example of information processing] A technique for calculating a matching score between sub-outfits, called set matching, is known. Set matching is, for example, a technique that uses deep learning, and estimates that the higher the matching score, the better the compatibility between the sub-outfits. Since the higher the matching score, the better the compatibility between the sub-outfits, it becomes possible to select the optimal combination from a large number of candidates when proposing a coordination. Using set matching makes it possible to quantitatively evaluate the compatibility between the sub-outfits.

[0017] In the following embodiment, information processing for providing highly harmonious information by applying the set matching technology will be described. In the following embodiment, a set that is the target of set matching will be referred to as a "target group" as appropriate. The target group is a target group that combines multiple targets. For example, the target group may be an item group called a query sub-code described in Patent Document 1, or may be an item group that combines multiple items owned (possessed) by a user (for example, multiple items that are estimated to be owned by the user from purchase data, multiple items that the user has registered as items owned by the user, etc.). In addition, for example, the target group may be an item group called a gallery sub-code described in Patent Document 1, or may be an item group that combines multiple items sold in a specific online shopping mall. The target group may be a set of item images (feature values ​​extracted from item images registered in various services such as electronic commerce services, item images photographed and registered by a user in various services such as a coordination service, etc.), or a set of item data (feature values ​​extracted from item data registered in various services such as electronic commerce services (item ID, category data, color data, pattern / design data, shape data, size data, material data, etc.), item data registered by a user in various services such as a coordination service, etc.). The target group may be an item group combining a plurality of items included in the posted information (e.g., a plurality of items estimated from tag data, etc.). Specifically, the target group may be an item group combining a plurality of items included in an image showing a coordination posted by a user in various services such as a coordination service, or an item group combining some of a plurality of items included in an image showing a coordination posted by a user.The target group may be a group of items that combines multiple item data such as description and tag data linked to an image showing a coordination posted by a user on various services such as a coordination service, or a group of items that combines some of multiple item data such as description and tag data linked to an image showing a coordination posted by a user. This makes it possible to select a group of items that are highly in harmony with all or some of the multiple items included in the coordination image posted by the user using set matching.

[0018] In the following embodiments, an "item" is a "multimodal object consisting of an item image alone or an item image and item attribute information (product category, product color, pattern / design, shape, size, material, etc.)." An "item" is converted into a single item feature vector by a feature extractor consisting of a convolutional neural network or a multilayer perceptron. Since coordinates, gallery items, and query items are objects consisting of multiple items, they are converted into multiple item feature vectors (hereinafter referred to as "item feature vector sets") by the above-mentioned feature conversion. The target group of the three-axis set matching model of the present application is composed of item feature vector sets.

[0019] Furthermore, in the following embodiments, the multiple objects in the object group are not limited to objects related to, for example, fashion, interior, etc., and may be any objects. Furthermore, the multiple objects in the object group do not have to be objects related to the same field. For example, the multiple objects in the object group may be objects related to fashion, interior, etc. For example, when set matching is related to meals, the multiple objects may be objects related to cooking or preparation (for example, ingredients, seasonings, cooking utensils, dishes, tableware, cutlery, small items (tablecloths, napkins, etc.), etc.). For example, the object group may be an ingredient group combining multiple ingredients, a seasoning group combining multiple seasonings, a dish group combining multiple dishes, or a tableware group combining multiple tableware, but is not limited to these. Also, when the set matching is related to meals, for example, matching may be performed on three axes: a food group combining multiple food ingredients, a seasoning group combining multiple seasonings, and a cooking utensil group combining multiple cooking utensils; a food group combining multiple food ingredients, a tableware group combining multiple tableware, and a cooking utensil group combining multiple cooking utensils; a food group combining multiple dishes, a tableware group combining multiple tableware, and a cutlery and small item group combining multiple cutlery and small items; a food group combining multiple dishes, a tableware and cutlery group combining multiple tableware and cutlery, and a small item group combining multiple small items; a food group combining multiple dishes, a tableware group, cutlery group, and small item group combining multiple tableware and cutlery, and a small item group combining multiple small items; a food group combining multiple dishes, a tableware group, cutlery group, and small item group combining multiple tableware, cutlery and small items, and a dining group combining multiple diners. Also, for example, when the set matching is related to a party, the multiple targets may be users who wish to participate in the party. For example, the multiple targets may be male members and female members.For example, the target group may be a male group made up of a combination of multiple male members (multiple users estimated based on age, occupation, hobbies, fashion preferences, etc.), or a female group made up of multiple female members (multiple users estimated based on age, occupation, hobbies, fashion preferences, etc.), but is not limited thereto. In this case, for example, a matching score may be calculated for a combination of a male group and a female group. In addition, when the set matching is related to a group date, for example, matching may be performed on three axes: a male group made up of multiple male members, a female group made up of multiple female members, and an event data group made up of multiple event data (such as a venue and a date); a male group made up of multiple male members, a female group made up of multiple female members, and a venue data group made up of multiple venue data (such as the venue location and the venue atmosphere, the first party venue, the second party venue, etc.); a male group made up of multiple male members, a female group made up of multiple female members, and a talk data group made up of multiple talk data (such as exciting talk data); or a male group made up of multiple male members, a female group made up of multiple female members, and a game data group made up of multiple game data (such as exciting game data). In addition, for example, when the set matching is related to a matching service, the multiple targets may be things related to fashion (such as fashion items) and user data (such as body data, makeup data, purchase data, etc.). For example, the target group may be an item group combining fashion items, or a user data group combining multiple user data of male users (or female users), but is not limited to these. In addition, when the set matching is related to a matching service, matching may be performed along three axes, for example, an item group combining multiple items, a male user data group combining multiple male user data, and a female user data group combining multiple female user data.

[0020] In the following embodiment, the user data may be any information as long as it is user data. The user data is a new axis that is a target of set matching along with one axis of the target group. The user data may be a set of user data that is a target of set matching, or may be a set of different types of user data. That is, the user data may be expressed as other targets or other target groups. Although details will be described later, the information processing device 100 may train a model so that the closer the combination of the input target group, user data, and additional information is to a combination (correct answer data) of a predetermined target group, predetermined user data, and predetermined additional information, the higher the matching score is calculated (output), and the information processing device 100 may train a model so that the better the compatibility between the input target group, user data, and additional information is, the higher the matching score is calculated (output). For example, the predetermined user data may be user data of a poster who posted a coordination image including a predetermined target group. And the user data may be user data of a user or the like. The user data may be, for example, physical data of the poster or user (height data, weight data, body type data, foot type data, hand type data, face type data, skin color data, skin condition data, etc.). For example, the user data of the poster or user may be physical data such as measurement data or user image of the poster or user. The user data may also be physical data of the poster or user estimated from measurement data, questionnaire data, purchase data, user image, etc. The user data may also be makeup data of the poster or user. For example, the user data may be makeup data such as makeup images posted by the user to various services such as a coordination service, or cosmetic data (item data) or cosmetic images (item images) linked to the makeup images and registered. The user data may also be makeup data related to makeup estimated from the makeup images, cosmetic data, or cosmetic images. The makeup data may also be makeup data for each part of the face. The user data may also be purchase data of the poster or user. For example, the user data may be items, item data, or item images purchased by the user in an electronic commerce service.In addition, this is the case where the wearer is the poster when the multiple objects are related to fashion, and when the wearer is not the poster, the predetermined user data may be the user data of the wearer in the coordinated image including the predetermined object group. In addition, the user data may be treated as a set of features so as to be able to cope with a case where the user data is partially missing. For example, in a case where body type data and face type data are required as user data, even if only the user's body type data exists, the matching score can be calculated as long as it is treated as a set of features of the user data. In addition, the user data may be, for example, an item group combining multiple items purchased by the user, or an item group combining multiple items handled by a predetermined electronic commerce service (such as an item group combining multiple items not purchased by the user).

[0021] In the following embodiment, the additional information may be any information as long as it is additional information for adding predetermined information regarding a plurality of targets. The additional information is a new axis that is a target of set matching together with one axis of the target group. The additional information may be a set of additional information that is a target of set matching, or a set of different types of additional information. That is, the additional information may be expressed as other targets or other target groups. Although details will be described later, the information processing device 100 may train a model so that the closer the combination of the input target group, user data, and additional information is to the combination (correct answer data) of a predetermined target group, predetermined user data, and predetermined additional information, the higher the matching score is calculated (output). For example, the predetermined additional information may be user data of another poster who posted a coordinated image including the target group. The user data may be, for example, physical data of another poster or user. Note that details of the user data are the same as those in the case of the above-mentioned user data being an axis, and therefore will not be described.

[0022] In addition, although details will be described later, the information processing device 100 may train a model so that the closer the input combination of the target group, user data, and additional information is to a combination of a highly rated target group, predetermined user data, and predetermined additional information, the higher the matching score calculated. For example, when a viewer who viewed a coordination image (an example of posted information) including a predetermined target group evaluates the coordination image, the predetermined additional information may be the user data (i.e., annotator data) of the viewer (evaluator). The evaluation of the coordination may be, for example, an evaluation of the coordination (a combination of items that are the predetermined target group) being "good" or "bad," or an evaluation of the coordination (items that are the predetermined target group) being "suitable" or "unsuitable" for the poster (or wearer) (predetermined user data). The annotator data may be attribute data such as the viewer's body data, makeup data, and purchase data, as well as the viewer's gender data, age data, residential address data, and occupation data. The annotator data may be preference data indicating the viewer's taste in fashion, knowledge data indicating the viewer's knowledge in fashion (or knowledge data indicating knowledge for each fashion style), or sensitivity data indicating the viewer's fashion sensitivity. The additional information may be user data (body data, makeup data, purchase data, attribute data, preference data, knowledge data, sensitivity data, etc.) of the user or a third party (the user's family, friends, lovers, dating partners, matching partners, etc.).

[0023] Further, although details will be described later, the information processing device 100 may train a model so that the closer the input combination of a target group, user data, and additional information is to a combination of a predetermined target group and predetermined user data, and the combination of the predetermined additional information that has been displayed together with the combination and has led to conversion, the higher the matching score is calculated (output). For example, if the predetermined additional information is an advertisement displayed while viewing a coordination image including a predetermined target group and has led to conversion, the data may be data related to the advertisement (hereinafter referred to as "advertisement data" as appropriate). The advertisement data may be, for example, advertiser data, content data, position data, size data, etc. Furthermore, the advertisement data may be, for example, advertisement data related to information estimated from the poster's coordination image (for example, the poster's (or wearer's) makeup, travel destination, etc.). In this case, the information processing device 100 may train a model so that the closer the input combination of a target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the higher the matching score is calculated (output).

[0024] In addition, although details will be described later, the information processing device 100 may train a model so that the closer the input combination of the object group, the user data, and the additional information is to a combination of a predetermined object group, the predetermined user data, and the predetermined additional information, the higher the matching score calculated (output). For example, the predetermined additional information may be data on the state of multiple objects. And the additional information may be data on the state of multiple objects. For example, when multiple objects are related to fashion, the data on the state of multiple objects may be data on the way multiple objects are dressed. The way of dressing may be data on the way of dressing, such as, for example, taking a shirt out of pants or wrapping a jacket around the waist. The way of dressing is important, and the impression is different depending on whether or not to roll up pants, and whether to put on or put on a jacket. For example, the way of wearing pants includes whether to roll up or not, and the way of wearing a jacket includes whether to put the sleeves through or put them on. For a hat, the brim can be facing forward, sideways, or backward, and whether to wear it parallel to the ground or at an angle. For an outerwear, the way of wearing it includes whether to wear it in a standard way like a business shirt, whether to wear it without putting the sleeves through, whether to wrap it around the waist, whether to tie it diagonally from the shoulder, whether to tuck the shirt in, or to tie it in the standard way and then tie it in front. For an innerwear, the way of tucking in the shirt or not, and for bottoms, whether to tuck bottoms into boots or not. Furthermore, the data on the state of the multiple objects is not limited to the case where the multiple objects are related to fashion, and may be data on any kind of information (what state of what kind of objects). For example, when the multiple objects are related to interior design, the data on the state of the multiple objects may be data on how to place the multiple objects.

[0025] 2 is a diagram showing an example of information processing by the information processing system 1 according to the embodiment. In FIG. 2, a proposal is made to a user U1 via the information display device 10.

[0026] The information processing device 100 generates a model that calculates a matching score based on set matching (step S101). Specifically, when the information processing device 100 receives a combination of a target group, user data, and additional information, it generates a model that outputs a matching score. The matching score indicates the degree of harmony of the combination of the target group, user data, and additional information. Here, the learning of the model will be described.

[0027] The information processing device 100 trains the model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data related to the predetermined target group, and predetermined additional information related to the predetermined target group, the better the compatibility between the input target group, user data, and additional information, and the higher the matching score. For example, when the predetermined additional information is user data of another poster (or wearer), the information processing device 100 may train the model so that the closer the combination is to a combination of the poster's coordinated image (one example of a predetermined target group), the poster's user data, and the user data of another poster (e.g., another poster who posted a coordinated image identical or similar to the poster's coordinated image), the higher the matching score. By training the model with a combination of the target group, user data, and additional information in this way, the information processing device 100 can, for example, propose a coordination that matches the physical characteristics (e.g., body type, etc.) of the user U1 and other physical characteristics that match the coordination (for example, determine a target group and user data that are compatible with the physical data of the user U1). The information processing device 100 can also propose other physical features that match the physical features and coordination of the user U1 (for example, determine user data that is compatible with the physical data of the user U1 and a target group). The information processing device 100 can also propose various physical features that match the coordination of the user U1 (for example, determine physical data and user data that are compatible with a target group). The information processing device 100 may also propose the determined physical features in a proposal manner that allows the determined physical features to be set as target physical features in various services such as a body management service. The information processing device 100 can also propose coordination that matches the makeup of the user U1 and physical features that match the coordination (for example, determine a target group and user data that are compatible with the makeup data of the user U1). The information processing device 100 can also propose coordination that matches the makeup and physical features of the user U1 (for example, determine a target group that is compatible with the makeup data and user data of the user U1).The information processing device 100 can also propose makeup that matches the coordination and physical features of the user U1 (for example, determine makeup data that is compatible with a target group and user data). The information processing device 100 can also propose makeup and physical features that match the coordination of the user U1 (for example, determine makeup data and user data that are compatible with a target group). The information processing device 100 may propose the determined makeup in a proposal form that allows the user to virtually try on the makeup in various services such as an electronic commerce service or a coordination service. The information processing device 100 can also propose coordination according to the purchasing tendency of the user U1 and physical features that match the coordination (for example, determine a target group and user data that are compatible with the purchasing data). That is, the information processing device 100 can propose coordination based on the correlation between the purchasing tendency and the posting tendency of the coordination image. The information processing device 100 can also propose coordination and items according to the posting tendency of the coordination image of the user U1 (for example, determine purchasing data that is compatible with the target group and user data). The information processing device 100 may propose the determined coordination or items in a proposal form that allows the coordination or items to be purchased in various services such as an electronic commerce service, a coordination service, etc. The above-mentioned example of the proposal is merely an example and is not particularly limited.

[0028] The information processing device 100 may train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group with a high evaluation for the target group, predetermined user data for the predetermined target group with a high evaluation, and predetermined additional information for the predetermined target group with a high evaluation, the higher the matching score becomes. For example, when the predetermined additional information is user data of another poster, the information processing device 100 may train a model so that the closer the combination is to a combination of a coordinated image of a poster with a high ranking based on a questionnaire on whether it looks good (one example of a predetermined target group), the poster's user data, and user data of another poster (for example, another poster who posted a coordinated image that is the same as or similar to the coordinated image of a poster with a high ranking based on a questionnaire on whether it looks good, etc.). By taking the ranking into consideration in this way, the information processing device 100 can, for example, propose a coordination that matches the physical characteristics (for example, body type, etc.) of the user U1, taking the ranking into consideration, and other physical characteristics that match the coordination. The information processing device 100 can, for example, propose other physical characteristics that match the physical characteristics of the user U1 and the coordination, taking the ranking into consideration. Furthermore, the information processing device 100 can propose various physical features that match the coordination of the user U1, for example, taking into account the ranking. Furthermore, the information processing device 100 can propose a coordination that matches the makeup of the user U1, for example, taking into account the ranking, and physical features that match the coordination. Furthermore, the information processing device 100 can propose a coordination that matches the makeup and physical features of the user U1, for example, taking into account the ranking. Furthermore, the information processing device 100 can propose makeup that matches the coordination and physical features of the user U1, for example, taking into account the ranking. Furthermore, the information processing device 100 can propose makeup that matches the coordination and physical features of the user U1, for example, taking into account the ranking. The information processing device 100 can propose makeup that matches the coordination and physical features of the user U1, for example, taking into account the ranking. The examples of such proposals are merely examples and are not particularly limited.The information processing device 100 may provide the evaluator with a questionnaire on whether the outfit suits them in various services such as an electronic commerce service or a coordination service. The questionnaire on whether the outfit suits them may be, for example, a questionnaire that asks the evaluator to rate coordination images of multiple posters as "good" or "bad," or a questionnaire that asks the evaluator to rate coordination images of multiple posters (coordination images including the posters (or wearers)) as "suitable" or "unsuitable." In addition, the ranking based on the questionnaire on whether the outfit suits them is such that the more highly rated outfits are, such as "good" or "suitable," the higher the ranking.

[0029] Also, for example, when the predetermined additional information is annotator data of an evaluator, the information processing device 100 may train a model so that the closer the combination is between the poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and the annotator data of an evaluator who has a high evaluation on the coordinated image (for example, an evaluator who has evaluated that the coordinated image suits the poster), the higher the matching score. Also, as in the case where the predetermined additional information is the user data of another poster, the information processing device 100 may train a model taking into consideration the ranking. In this way, the information processing device 100 can propose a coordination (for example, a coordination for a date) that a third party such as the lover of the user U1 evaluates as suiting the user U1 (determine a target group that is compatible with the user data of the third party and the user data of the user U1). Furthermore, the information processing device 100 can, for example, in a matching service, propose a matching partner recommended to the user U1 and a coordination that the matching partner evaluates as suiting the user U1 (for example, determine the user data of a third party that is compatible with the user data of the user U1 and a target group). Such an example of a proposal is an example and need not be particularly limited. Furthermore, the information processing device 100 may perform set matching on three axes: the target group, another target group (a target group combining multiple targets different from the target group), and additional information, instead of the three axes: the target group, the user data, and the additional information. In this case, the information processing device 100 may perform set matching on three axes: the coordinated image of the poster (or wearer) (an example of a combination of a predetermined target group and a predetermined other target group) and the annotator data, with the target group and the other target group each being a new axis. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordination image and the annotator data of an evaluator who gave a high rating to the coordination image (e.g., an evaluator who rated the coordination as good), the higher the matching score.Thereby, the information processing device 100 can propose, for example, a coordination or sub-coordination that is evaluated as good by a third party such as the user U1's family, friend, lover, blind date partner, or match partner when worn by the user U1 (determine a first target group and a second target group that are compatible with the user data of the third party, or determine the remaining target group that is compatible with the user data of the third party and any of the target groups). In addition, the information processing device 100 can propose, for example, a coordination or sub-coordination that is appreciated as a gift to a third party such as the user U1's family, lover, friend, blind date partner, or match partner (determine a first target group and a second target group that are compatible with the user data of the third party, or determine the remaining target group that is compatible with the user data of the third party and any of the target groups). Such an example of the proposal is merely an example and need not be particularly limited.

[0030] Also, for example, when the predetermined additional information is advertisement data, the information processing device 100 may train the model so that the closer the combination of the poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and advertisement data that is displayed together with the coordinated image and has led to a conversion (for example, leading to a click, leading to a product purchase or product reservation, leading to a service application or service reservation, etc.), the higher the matching score. Also, similar to the case where the predetermined additional information is the user data of another poster, the information processing device 100 may train the model taking into consideration ranking. Also, the information processing device 100 may train the model based on a combination with advertisement data related to information estimated from the poster's coordinated image. As a result, the information processing device 100 can, for example, propose a recommended advertisement together with a coordinate that matches the physical characteristics of the user U1 (for example, determine a target group and advertisement data that are compatible with the physical data of the user U1). Furthermore, when the user U1 logs in to a predetermined posting site, the information processing device 100 can provide a service of proposing, as a top screen, a coordination that matches the registered body type of the user and an advertisement that goes well with the coordination (for example, determining a target group and advertisement data that go well with the user data of the user U1). Furthermore, the information processing device 100 can provide a service of proposing an advertisement that goes well with the registered body type of the user U1 and the posted coordination (determining advertisement data that goes well with the user data and the target group of the user U1) as a screen such as a posting screen or a profile screen of the user U1 on a predetermined posting site. Such an example of a proposal is an example and is not particularly limited. Furthermore, the information processing device 100 may perform set matching on three axes, namely, the target group, another target group, and additional information, instead of the three axes of the target group, the user data, and the additional information. In this case, the information processing device 100 may perform set matching on three axes, namely, the target group and the other target group, respectively, as new axes, namely, the poster's coordinated image (an example of a combination of a predetermined target group and a predetermined other target group) and advertisement data.For example, the information processing device 100 may train a model so that the closer the combination is between the poster's coordinated image and the advertising data that is displayed together with the coordinated image and that led to conversion, the higher the matching score. Furthermore, the information processing device 100 performs set matching on three axes, the target group, other target groups, and additional information, so that the information processing device 100 can, for example, propose recommended advertisements along with recommended coordination (for example, determine a first target group, a second target group, and advertising data that are compatible). Furthermore, when the user U1 purchases items A and B, the information processing device 100 can provide a service of proposing, as a post-purchase screen, a coordination consisting of items A and B and items C and D that are compatible with items A and B, respectively, and an advertisement that is compatible with the coordination (for example, determine advertising data that is compatible with the first target group and the second target group).

[0031] The information processing device 100 may also train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information, and the higher the matching score. For example, when the predetermined additional information is data on the state of a plurality of targets (hereinafter referred to as "state data" as appropriate), the information processing device 100 may train a model so that the closer the combination is to a poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and the state data related to the coordinated image, the higher the matching score. For example, the information processing device 100 may take into account the way of dressing by performing set matching on three axes: the poster's coordinated image, the poster's user data, and the dressing data related to the poster's dressing. The condition data may be, for example, description or tag data (for example, description or tag data explaining the condition such as the way of dressing) registered by a user in association with a coordination image in various services such as a coordination service, data on the condition of multiple objects estimated from the coordination image, or fashion data from a predetermined fashion site. In addition, as in the case where the predetermined additional information is the user data of another poster, the information processing device 100 may train a model taking into consideration the ranking. In this way, the information processing device 100 can propose a way of dressing that matches the coordination of the user U1 and takes into consideration the physical characteristics such as the body type of the user U1 by performing set matching on three axes of the poster's coordination image, the poster's user data, and the dressing data related to the poster's dressing (determine condition data that is compatible with the target group and the user data of the user U1). In addition, the information processing device 100 can perform set matching along three axes: the poster's coordination image, the poster's user data, and the style data related to the poster's style, thereby proposing coordination and style that take into account the physical characteristics of user U1 (determining a target group and condition data that are compatible with the user data of user U1).Furthermore, the information processing device 100 can propose a coordination that takes into account the physical characteristics of the user U1 and matches the state of multiple targets designated by the user U1 (determine a target group that is compatible with the user data and state data of the user U1) by performing set matching on three axes: the poster's coordinated image, the poster's user data, and the poster's fashion data. Such an example of a proposal is an example and need not be particularly limited. Furthermore, in cases where the state of multiple targets is related to fashion, the information processing device 100 is not limited to a case where the model is learned using the state data as predetermined additional information, but may include the state data in the item data to learn the model. When the information processing device 100 learns a model by including the state data in the item data (including the state data in a predetermined target group), for example, the model may be learned by adding the fashion to the item feature (the feature of the target group). Furthermore, the information processing device 100 may perform set matching on three axes: the target group, another target group, and additional information, instead of the target group, user data, and additional information. At this time, the information processing device 100 may perform set matching on three axes, for example, the coordinated image of the poster (one example of a combination of a predetermined target group and a predetermined other target group) and the fashion data, with each of the target group and the other target group as a new axis. For example, the information processing device 100 may train a model so that the closer the combination of the coordinated image of the poster and the status data related to the poster's fashion, the higher the matching score. Also, the information processing device 100 may train a model so that the closer the combination of the coordinated image of the poster and the fashion data, the higher the matching score. Thereby, for example, the information processing device 100 can propose a coordination that matches the status of multiple targets specified by the user U1 (for example, determine a first target group and a second target group that are compatible with the status data). Also, the information processing device 100 can propose a status of multiple targets that matches the coordination of the user U1 (for example, determine status data that is compatible with the first target group and the second target group). Such an example of a proposal is an example and is not particularly limited.

[0032] The information processing device 100 may also train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information and the higher the matching score. For example, when the predetermined additional information is data related to a restaurant (hereinafter referred to as "restaurant data" as appropriate), the information processing device 100 may train a model so that the closer the combination is to a poster's coordinated image (an example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and restaurant data highly related to the coordinated image, the higher the matching score. The restaurant data is restaurant data such as the type of restaurant (Japanese restaurant, Chinese restaurant, Western food restaurant, etc.), the cuisine of the restaurant, the location of the restaurant, and the time of visiting the restaurant. For example, the restaurant data may be data related to the restaurant estimated from the description or tag data (for example, description or tag data explaining the visited place) registered by the user in association with the coordinated image in various services such as a coordinate service or the background of the coordinated image, or reservation data from a predetermined reservation site or the like, and data related to the restaurant estimated from the reservation data. In addition, as in the case where the predetermined additional information is the user data of another poster, the information processing device 100 may train a model taking into account the ranking. This allows the information processing device 100 to propose a restaurant that matches the coordination of the user U1 and takes into account the characteristics of the user U1 (determine the target group and restaurant data that are compatible with the user data of the user U1). In addition, the information processing device 100 can propose a coordination and a restaurant that take into account the characteristics of the user U1 (determine the target group and restaurant data that are compatible with the user data of the user U1). In addition, the information processing device 100 can propose a coordination that takes into account the characteristics of the user U1 and suits the restaurant that the user U1 plans to go to (determine a target group that is compatible with the user data and restaurant data of the user U1). Such an example of a proposal is merely an example and is not particularly limited.The information processing device 100 may perform set matching not on the three axes of the target group, user data, and additional information, but on the three axes of the target group, other target groups, and additional information. In this case, the information processing device 100 may perform set matching on the three axes of the poster's coordinated image (one example of a combination of a predetermined target group and a predetermined other target group) and restaurant data, with the target group and the other target groups as new axes. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordinated image (one example of a combination of a predetermined target group and a predetermined other target group) and restaurant data highly related to the coordinated image, the higher the matching score. This allows the information processing device 100 to propose, for example, a coordinate that matches the restaurant that the user U1 plans to go to (for example, determine the first target group and the second target group that are compatible with the restaurant data). Also, the information processing device 100 may propose, for example, a restaurant that matches the coordinate that the user U1 wants to wear to the restaurant (for example, determine the restaurant data that is compatible with the first target group and the second target group). Furthermore, the information processing device 100 can, for example, propose a sub-coordinate and a restaurant that match the sub-coordinate that the user U1 wants to wear to a restaurant (for example, determine the remaining target group and restaurant data that are compatible with any of the target groups). Furthermore, the information processing device 100 can, for example, propose a restaurant that the user U1 plans to go to and a sub-coordinate that matches the sub-coordinate that the user U1 wants to wear to the restaurant (for example, determine the restaurant data and the remaining target group that are compatible with any of the target groups). Such an example of a proposal is an example and may not be particularly limited. Furthermore, a restaurant is an example, and may be various places such as a shop, a cafe, a bar, a hotel, etc. In this case, the restaurant data may be called visited data.

[0033] The information processing device 100 may also train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information, and the higher the matching score. For example, when the predetermined additional information is data related to interiors (hereinafter referred to as "interior data" as appropriate), the information processing device 100 may train a model so that the closer the combination is to a combination of a poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and interior data highly related to the coordinated image, the higher the matching score. The interior data is interior data such as the type of interior, the color of the interior, the shape of the interior, and the size of the interior, and may be, for example, a description or tag data (for example, interior data describing the background interior, etc.) linked to a coordinated image and registered by a user in various services such as a coordination service, or data related to the interior estimated from the background of the coordinated image, or may be purchase data related to the interior on a predetermined shopping site, or data related to the interior estimated from the purchase data. Also, similarly to the case where the predetermined additional information is the user data of another contributor, the information processing device 100 may train the model taking into consideration the ranking. This allows the information processing device 100 to propose an interior that matches the coordination of the user U1 and takes into consideration the characteristics of the user U1 (determine interior data that is compatible with the target group and the user data of the user U1). Also, the information processing device 100 can propose a coordination and an interior that takes into consideration the characteristics of the user U1 (determine a target group and interior data that are compatible with the user data of the user U1). Also, the information processing device 100 can propose a coordination that matches the interior owned by the user U1, taking into consideration the characteristics of the user U1 (determine a target group that is compatible with the user data and interior data of the user U1). Such an example of a proposal is merely an example and need not be particularly limited.The information processing device 100 may perform set matching not on the three axes of the target group, user data, and additional information, but on the three axes of the target group, other target groups, and additional information. In this case, the information processing device 100 may perform set matching on the three axes of the poster's coordinated image (one example of a combination of a predetermined target group and a predetermined other target group) and interior data, with the target group and the other target groups as new axes. For example, the information processing device 100 may train a model so that the closer the combination is between the poster's coordinated image (one example of a combination of a predetermined target group and a predetermined other target group) and interior data that is highly related to the coordinated image, the higher the matching score. This allows the information processing device 100 to propose, for example, a coordination that matches the interior owned by the user U1 (for example, determine the first target group and the second target group that are compatible with the interior data). The information processing device 100 may also propose, for example, an interior that matches the coordination owned by the user U1 (for example, determine the interior data that is compatible with the first target group and the second target group). Furthermore, the information processing device 100 can, for example, propose sub-coordination and interior that matches the sub-coordination of the user U1 (for example, determine the remaining object group and interior data that are compatible with any of the object groups). Furthermore, the information processing device 100 can, for example, propose interior owned by the user U1 and sub-coordination that matches the sub-coordination of the user U1 (for example, determine the interior data and the remaining object group that are compatible with any of the object groups). Such an example of proposal is merely an example and need not be particularly limited.

[0034] Furthermore, the information processing device 100 may perform set matching not on the three axes of the target group, the user data, and the additional information, but on the three axes of the target group, the other target group, and the additional information. For example, when the predetermined additional information is ideal data of the poster's coordinated image (or data showing the difference between the usual coordinated image and the ideal coordinated image), the information processing device 100 may train a model so that the closer the combination of the poster's coordinated image, the poster's user data, and the poster's ideal data is, the higher the matching score will be. Furthermore, the information processing device 100 may use a conventional technique of VSE (Visual-Semantic Embedding) to calculate a feature amount (vector) from the usual coordinated image to the ideal coordinated image, and use the feature amount for set matching. Similarly to the case where the predetermined additional information is the poster's user data, the information processing device 100 may train a model taking into consideration the ranking. As a result, the information processing device 100 can propose, for example, an ideal coordinate that matches the physical characteristics of the user U1. Furthermore, the information processing device 100 can resolve a user's concern that, even if a user wants to try a different outfit from usual, the user is reluctant to try an outfit that is too far removed from usual outfits.

[0035] The information processing device 100 may also train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information, and the higher the matching score. For example, when the predetermined additional information is data related to weather (hereinafter referred to as "weather data" as appropriate), the information processing device 100 may train a model so that the closer the combination is to a combination of a poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and weather data related to the coordinated image (for example, weather data for the date, time, and place when the coordinated image was taken, etc.). The weather data is weather data such as weather, temperature, and probability of precipitation, and may be, for example, a description or tag data (for example, a description or tag data explaining the weather when worn, etc.) linked to a coordinated image and registered by a user in various services such as a coordinate service, or data related to the weather estimated from the background of the coordinated image, or weather data from a predetermined weather forecast site, etc. Also, similarly to the case where the predetermined additional information is the user data of another poster, the information processing device 100 may train the model taking into consideration the ranking. As a result, the information processing device 100 can, for example, propose user data and weather information that match the coordination of the user U1 (for example, determine user data and weather data that are compatible with a target group). Also, the information processing device 100 can propose weather information that matches the coordination of the user U1 and takes into consideration the characteristics of the user U1 (determine weather data that is compatible with the target group and the user data of the user U1). Also, the information processing device 100 can propose coordination and weather information that takes into consideration the characteristics of the user U1 (determine a target group and weather data that are compatible with the user data of the user U1). Also, the information processing device 100 can propose coordination that matches the weather forecast for the date, time, and location of the plan of the user U1 taking into consideration the characteristics of the user U1 (determine a target group that is compatible with the user data and weather data of the user U1).Such a proposal example is merely an example and may not be particularly limited. In addition, the information processing device 100 may perform set matching not on the three axes of the target group, user data, and additional information, but on the three axes of the target group, other target groups, and additional information. In this case, the information processing device 100 may perform set matching on the three axes of the poster's coordinated image (an example of a combination of a predetermined target group and a predetermined other target group) and weather data, with the target group and the other target groups as new axes, for example. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordinated image (an example of a combination of a predetermined target group and a predetermined other target group) and the weather data related to the coordinated image is, the higher the matching score will be. As a result, the information processing device 100 can, for example, propose a coordination that matches the weather forecast for the date, time, and location of the user U1's plan (for example, determine the first target group and the second target group that are compatible with the weather data). The information processing device 100 can also propose weather information that matches the coordination of the user U1 (for example, determine weather data that goes well with the first target group and the second target group). The information processing device 100 can also propose sub-outfits and weather information that match the sub-coordination of the user U1 (for example, determine the remaining target group and weather data that go well with any of the target groups). The information processing device 100 can also propose weather forecasts for the date, time, and location of the plan of the user U1 and sub-outfits that match the sub-coordination of the user U1 (for example, determine weather data and the remaining target group that goes well with any of the target groups). Such examples of suggestions are merely examples and are not particularly limited.

[0036] Furthermore, the information processing device 100 may train a model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information and the higher the matching score. For example, when the predetermined additional information is data related to TPO (Time Place Occasion) (hereinafter referred to as "TPO data" as appropriate), the information processing device 100 may train a model so that the closer the combination is to a combination of a poster's coordinated image (an example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and the TPO data related to the coordinated image, the higher the matching score. The TPO data is TPO data such as time, place, and scene, and may be, for example, data related to the TPO estimated from a description or tag data (for example, description or tag data explaining the TPO when worn, etc.) registered by a user in various services such as a coordination service in association with a coordination image, or a background of a coordination image, or reservation data such as a predetermined travel site, or TPO data estimated from the reservation data, or schedule data such as a predetermined schedule application, or TPO data estimated from the schedule data. In addition, similar to the case where the predetermined additional information is user data of another poster, the information processing device 100 may train a model taking into consideration the ranking. Thereby, the information processing device 100 can, for example, propose user data and TPO information that match the coordination of the user U1 (for example, determine user data and TPO data that are compatible with a target group). In addition, the information processing device 100 can propose TPO information that matches the coordination of the user U1 and takes into consideration the characteristics of the user U1 (determine interior data that is compatible with the target group and the user data of the user U1). Furthermore, the information processing device 100 can propose coordination and TPO information that take into account the characteristics of the user U1 (determine a target group and TPO data that are compatible with the user data of the user U1).Furthermore, the information processing device 100 can propose a coordination that takes into account the characteristics of the user U1 and matches the planned TPO of the user U1 (determine a target group that is compatible with the user data and TPO data of the user U1). Such an example of a proposal is merely an example and need not be particularly limited. Furthermore, the information processing device 100 may perform set matching on three axes, that is, the target group, other target groups, and additional information, instead of the three axes, that is, the target group, user data, and additional information. In this case, the information processing device 100 may perform set matching on three axes, that is, the poster's coordinated image (an example of a combination of a predetermined target group and a predetermined other target group) and the TPO data, with the target group and the other target groups each being a new axis. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordinated image (an example of a combination of a predetermined target group and a predetermined other target group) and the TPO data related to the coordinated image is, the higher the matching score will be. Thereby, the information processing device 100 can propose, for example, a coordination that matches the planned TPO of the user U1 (for example, determine the first target group and the second target group that are compatible with the TPO data). The information processing device 100 can also propose, for example, TPO information that matches the coordination of the user U1 (for example, determine the TPO data that is compatible with the first target group and the second target group). The information processing device 100 can also propose, for example, a sub-coordinate and TPO information that matches the sub-coordinate of the user U1 (for example, determine the remaining target group and TPO data that are compatible with any of the target groups). The information processing device 100 can also propose, for example, a sub-coordinate that matches the planned TPO of the user U1 and the sub-coordinate of the user U1 (for example, determine the remaining target group that is compatible with any of the target groups with the TPO data). Such an example of a proposal is merely an example and is not particularly limited.

[0037] The information processing device 100 may also train the model so that the closer the combination of the input target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the better the compatibility between the input target group, user data, and additional information, and the higher the matching score. For example, when the predetermined additional information is data related to music (hereinafter referred to as "music data" as appropriate), the information processing device 100 may train the model so that the closer the combination is to a poster's coordinated image (one example of a predetermined target group), the poster's (or wearer's) user data related to the coordinated image, and music data highly related to the coordinated image (for example, music data that matches the image of the coordinated image or the poster's favorite music data, etc.), the higher the matching score. The music data may be music data such as music genres and artists, and may be, for example, explanatory text and tag data (e.g., explanatory text and tag data explaining music that matches the image of the coordinated image) linked to the coordinated image and registered, music data (e.g., music data as BGM selected when posting the coordinated image), music estimated from the background of the coordinated image (e.g., if the background is the sea, the music genre and artist that matches the sea), or music data from a predetermined music site. In addition, as in the case where the predetermined additional information is user data of other posters, the information processing device 100 may train a model taking into account rankings. Thereby, the information processing device 100 can, for example, propose a coordination that matches the music that the user U1 wants to listen to, taking into account the characteristics of the user U1 (for example, determine a target group that is compatible with the user data and music data of the user U1). In addition, the information processing device 100 can, for example, propose user data and music that match the coordination of the user U1 (for example, determine user data and music data that are compatible with the target group). Furthermore, the information processing device 100 can propose, for example, music that the user U1 wants to listen to and user data that matches the coordination of the user U1 (for example, determine user data that is compatible with the music data and a target group).In addition, the information processing device 100 can propose coordination and music that take into account the characteristics of the user U1 (determine a target group and music data that are compatible with the user data of the user U1). In addition, the information processing device 100 can propose music that matches the coordination of the user U1 and takes into account the characteristics of the user U1 (determine music data that is compatible with the target group and the user data of the user U1). Such an example of a proposal is an example and is not particularly limited. In addition, the information processing device 100 may perform set matching on three axes, namely, the target group, other target groups, and additional information, instead of the three axes of the target group, user data, and additional information. In this case, the information processing device 100 may perform set matching on three axes, namely, the poster's coordination image (an example of a combination of a predetermined target group and a predetermined other target group) and music data, with each of the target group and the other target group as a new axis. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordination image (an example of a combination of a predetermined target group and a predetermined other target group) and music data that is highly related to the coordination image, the higher the matching score. Thereby, the information processing device 100 can propose, for example, a coordination that matches the music that the user U1 wants to listen to (for example, determine the first target group and the second target group that are compatible with the music data). In addition, the information processing device 100 can propose, for example, music that matches the coordination of the user U1 (for example, determine the music data that is compatible with the first target group and the second target group). This makes it easier to select background music when posting a coordination image, and can lead to promoting the purchase of music on a predetermined music site. In addition, the information processing device 100 can propose, for example, a sub-outfit and music that match the sub-coordination of the user U1 (for example, determine the remaining target group and music data that are compatible with any of the target groups). In addition, the information processing device 100 can propose, for example, music that the user U1 wants to listen to and a sub-outfit that matches the sub-coordination of the user U1 (for example, determine the remaining target group that is compatible with the music data and any of the target groups). Such an example of a proposal is one example and is not particularly limited.Furthermore, the information processing device 100 may perform set matching along three axes: the poster's coordinated image, music data, and interior data. For example, the information processing device 100 may train a model so that the closer the combination of the poster's coordinated image, interior data, and music data highly related to the coordinated image, the higher the matching score. This allows the information processing device 100 to suggest music that matches the coordination of the user U1 and takes into account the characteristics of the interior (determine the target group and music data that is compatible with the interior). Furthermore, the information processing device 100 can suggest coordination that matches the music that the user U1 wants to listen to, taking into account the characteristics of the interior (determine the target group that is compatible with the interior data and music data).

[0038] The learning of the model has been described above. The model may be a model in which the target group, the user data, and the additional information are set as three variables and the matching score is calculated by varying all three variables, or may be a model in which the matching score is calculated by fixing at least one variable. Details of the calculation of the matching score when all three variables are varied and when at least one variable is fixed will be described later.

[0039] Returning now to the explanation of FIG. 2, the information processing device 100 acquires information on the combination of the target group, the user data, and the additional information (step S102). The target group, the user data, and the additional information are the targets of the calculation of the matching score. The target group, the user data, and the additional information may be determined in any manner. For example, the target group, the user data, and the additional information may be determined from item images, etc., may be determined from item data, etc., may be determined from user data of user U1, etc., may be determined from purchase data of user U1, or may be determined from coordinated images, item images, item data, etc. of user U1.

[0040] When the information processing device 100 acquires information on the combination of the object group, the user data, and the additional information, the information processing device 100 inputs the information on the combination of the object group, the user data, and the additional information into a predetermined model to calculate a matching score (step S103). This matching score indicates the degree of harmony of the acquired combination of the object group, the user data, and the additional information. The information processing device 100 acquires information on the combination of a plurality of object groups, the user data, and the additional information, and inputs the information on the combination into a predetermined model to calculate a matching score for each combination. Then, the information processing device 100 provides information on the combination of the object group, the user data, and the additional information that satisfies a predetermined condition by comparing the matching scores calculated for each combination (step S104).

[0041] In the above embodiment, the object group is a name indicating a group of a combination of multiple objects. The object group itself may be a variable as a group, or each object included in the object group may be a variable. In the latter case, one or more of the objects included in the object group may be a variable. Since the object group is a combination of multiple objects, processing may be performed taking into account multiple variables. In addition, providing information on the combination of the object group, user data, and additional information is not limited to providing all information on the object group, user data, and additional information, and includes, for example, providing only information on the object group. In addition, it includes providing only two pieces of information, information on the object group and information on the user data, providing only two pieces of information, information on the object group and information on the additional information, and providing only two pieces of information, information on the user data and information on the additional information. In other words, the information provided by the information processing device 100 is not limited to information that combines all information on the object group, user data, and additional information, and may be information on only the object group (which may be only user data or only additional information) in the combination of the object group, user data, and additional information. For example, when the information processing device 100 receives a designation of an outfit from a user, the information processing device 100 may only suggest additional information that matches the user's body type and the outfit.

[0042] 3. Configuration of Information Display Device Next, the configuration of the information display device 10 according to the embodiment will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the configuration of the information display device 10 according to the embodiment. As shown in Fig. 3, the information display device 10 has a communication unit 11, an input unit 12, an output unit 13, and a control unit 14.

[0043] (Communications Department 11) The communication unit 11 is realized by, for example, a network interface card (NIC) etc. The communication unit 11 is connected to a predetermined network N by wire or wirelessly, and transmits and receives information to and from the information processing device 100 etc. via the predetermined network N.

[0044] (Input section 12) The input unit 12 accepts various operations from a user. In the example shown in Fig. 2, the input unit 12 accepts various operations from a user U1. For example, the input unit 12 may accept various operations from a user via a display surface using a touch panel function. The input unit 12 may also accept various operations from buttons provided on the information display device 10, or a keyboard or mouse connected to the information display device 10.

[0045] (Output section 13) The output unit 13 is a display screen of a tablet terminal or the like realized by, for example, a liquid crystal display or an organic EL (Electro-Luminescence) display, and is a display device for displaying various information. For example, the output unit 13 displays information provided by the information processing device 100.

[0046] (Control unit 14) The control unit 14 is, for example, a controller, and is realized by a CPU (Central Processing Unit), MPU (Micro Processing Unit), or the like executing various programs stored in a storage device inside the information display device 10 using a RAM (Random Access Memory) as a working area. For example, the various programs include application programs installed in the information display device 10. For example, the various programs include application programs that display matching results obtained by set matching. The control unit 14 is also realized by an integrated circuit, for example, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0047] As shown in FIG. 3, the control unit 14 has a receiving unit 141 and a transmitting unit 142, and realizes or executes the information processing operations described below.

[0048] (Receiving unit 141) The receiving unit 141 receives various information from an external information processing device. For example, the receiving unit 141 receives various information from another information processing device such as the information processing device 100. For example, the receiving unit 141 receives information indicating a matching result by set matching. For example, the receiving unit 141 receives information regarding a combination of a target group, user data, and additional information that satisfy a predetermined condition.

[0049] (Transmitter 142) The transmission unit 142 transmits various information to an external information processing device. For example, the transmission unit 142 transmits various information to another information processing device such as the information processing device 100. For example, the transmission unit 142 transmits operation information performed by a user to execute set matching. For example, when a user designates multiple targets as candidates for a target group, when a user designates user data, or when a user designates additional information, the transmission unit 142 transmits the user's designation information together with the user's operation information.

[0050] 4. Configuration of Information Processing Device Next, the configuration of the information processing device 100 according to the embodiment will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the configuration of the information processing device 100 according to the embodiment. As shown in Fig. 4, the information processing device 100 has a communication unit 110, a storage unit 120, and a control unit 130. Note that the information processing device 100 may have an input unit (e.g., a keyboard, a mouse, etc.) that receives various operations from an administrator of the information processing device 100, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

[0051] (Communication unit 110) The communication unit 110 is realized by, for example, a NIC etc. The communication unit 110 is connected to a network N by wire or wirelessly, and transmits and receives information to and from the information display device 10 etc. via the network N.

[0052] (Storage unit 120) The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in FIG. 4, the storage unit 120 has a posted information storage unit 121, a model information storage unit 122, and a user information storage unit 123.

[0053] The posted information storage unit 121 stores information related to the content posted by the poster, linked to the content posted by the poster. The posted information storage unit 121 is used, for example, for generating and learning a model that calculates a matching score based on set matching. FIG. 5 shows an example of the posted information storage unit 121 according to the embodiment. As shown in FIG. 5, the posted information storage unit 121 has items such as "content ID," "contributor ID," "posting date and time," "coordinate image," "user data," "annotator data," and "advertising data."

[0054] "Content ID" indicates identification information for identifying content posted by a poster (for example, content posted by a poster on a specific site including a social networking service (SNS)). "Poster ID" indicates identification information for identifying the poster of the content. "Posting date and time" indicates the date and time the content was posted. "Coordinate image" indicates a coordinate image included in the content. Multiple objects in this coordinate image correspond to an object group. In the example shown in Figure 5, conceptual information such as "Coordinate image #1" and "Coordinate image #2" is stored in "Coordinate image", but in reality, image data, etc. are stored. In addition, for example, a URL where the image data is located, or a file path name indicating the storage location, etc. may be stored in "Coordinate image".

[0055] "User data" indicates user data of the poster (or wearer). In the example shown in FIG. 5, conceptual information such as "user data #1" and "user data #2" is stored in "user data", but in reality, physical data and makeup data of the poster (or wearer) are stored. "Annotator data" indicates annotator data related to a coordinated image (for example, a coordinated image of an coordinated image). In the example shown in FIG. 5, conceptual information such as "annotator data #1" and "annotator data #2" is stored in "annotator data", but in reality, user data of an evaluator (physical data, makeup data, etc.) is stored. In addition, data of a combination of an object evaluated by an evaluator and an evaluation may be stored. For example, when an evaluator specifies only a jacket from the coordinated image and evaluates that "the jacket does not look good on you, so you should change it", data of a combination of the jacket and an evaluation that "the jacket does not look good on you, so you should change it" may be stored. "Advertisement data" indicates advertisement data related to a candidate advertisement to be displayed in a predetermined area together with content, etc. In the example shown in FIG. 5, conceptual information such as "advertising data #1" and "advertising data #2" is stored in "advertising data," but in reality, advertiser data, content data, position data, size data, etc. are stored.

[0056] The model information storage unit 122 stores information about a model generated to calculate a matching score based on set matching. An example of the model information storage unit 122 according to the embodiment is shown in Fig. 6. As shown in Fig. 6, the model information storage unit 122 has items such as "model ID" and "model data".

[0057] "Model ID" indicates identification information for identifying a model. "Model data" indicates model data. In the example shown in FIG. 6, conceptual information such as "model data #1" and "model data #2" is stored in "model data", but in reality, data such as the weights of model variables (three variables: target group, user data, and additional information) is stored.

[0058] The user information storage unit 123 stores information about a user who performs an operation to execute set matching. The user information storage unit 123 is used, for example, to calculate a matching score using a model. When set matching is executed, a matching result by the set matching is provided to the user. Here, FIG. 7 shows an example of the user information storage unit 123 according to the embodiment. As shown in FIG. 7, the user information storage unit 123 has items such as "user ID" and "user data".

[0059] "User ID" indicates identification information for identifying a user. "User data" indicates the user data of the user. In the example shown in FIG. 7, conceptual information such as "User data #1" and "User data #2" is stored in "User data", but in reality, the user's body data, makeup data, etc. are stored.

[0060] (Control unit 130) The control unit 130 is a controller, and is realized, for example, by a CPU, an MPU, or the like, executing various programs stored in a storage device inside the information processing device 100 using a RAM as a working area. The control unit 130 is also a controller, and is realized, for example, by an integrated circuit such as an ASIC or an FPGA.

[0061] 4, the control unit 130 has an acquisition unit 131, a generation unit 132, a learning unit 133, a calculation unit 134, and a provision unit 135, and realizes or executes the information processing action described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 4, and may be other configurations as long as they perform the information processing described below.

[0062] (Acquisition part 131) The acquisition unit 131 acquires various pieces of information from an external information processing device. For example, the acquisition unit 131 acquires various pieces of information from another information processing device such as the information display device 10.

[0063] The acquisition unit 131 acquires various pieces of information from the storage unit 120. For example, the acquisition unit 131 acquires various pieces of information from the posted information storage unit 121, the model information storage unit 122, or the user information storage unit 123. In addition, the acquisition unit 131 stores the acquired various pieces of information in the storage unit 120. For example, the acquisition unit 131 stores the various pieces of information in the posted information storage unit 121, the model information storage unit 122, or the user information storage unit 123.

[0064] The acquisition unit 131 acquires information for generating a model that calculates a matching score based on set matching. The acquisition unit 131 also acquires information for training a model that calculates a matching score based on set matching.

[0065] The acquisition unit 131 acquires information for calculating a matching score using a model.

[0066] The acquisition unit 131 acquires information about a target group. The acquisition unit 131 also acquires a coordinated image of the target group. The acquisition unit 131 also acquires information about the target group from the coordinated image. The acquisition unit 131 also acquires user data. The acquisition unit 131 also acquires additional information. The acquisition unit 131 also acquires information about a combination of the target group, the user data, and the additional information.

[0067] The acquisition unit 131 acquires operation information of a user who has performed an operation to execute set matching. The acquisition unit 131 also acquires designation information of the user who has performed an operation to execute set matching.

[0068] (Generation unit 132) The generation unit 132 generates a model that calculates a matching score based on set matching. Specifically, the generation unit 132 generates a model that outputs a matching score when a combination of a target group, user data, and additional information is input.

[0069] Here, the model generated by the generation unit 132 may be a model in which the object group, the user data, and the additional information are set as three variables and a matching score is calculated by moving all three variables, or a model in which at least one variable is fixed and a matching score is calculated. The model generated by the generation unit 132 may also be a model in which at least one object in the object group is fixed and a matching score is calculated. Similarly, the model generated by the generation unit 132 may be a model in which at least one object in the user data is fixed and a matching score is calculated by fixing at least one object in the additional information. In addition, the model generated by the generation unit 132 may be a model corresponding to various variations in which a matching score is calculated by fixing one object and setting the other object as a variable.

[0070] The generating unit 132 generates a model that calculates a matching score by fixing, for example, two variables. For example, the generating unit 132 generates a model that outputs a matching score by fixing two variables, user data and additional information, and moving one variable of a target group. This matching score indicates the degree of harmony of a combination of a target group, user data, and additional information, and also indicates the degree of harmony of a combination of user data and additional information for each target group by moving one variable of the target group.

[0071] Also, for example, the generation unit 132 generates a model that outputs a matching score by fixing two variables, the object group and the additional information, and varying one variable, the user data. This matching score indicates the degree of harmony of the combination of the object group, the user data, and the additional information, and also indicates the degree of harmony of the combination of the object group and the additional information for each user data by varying one variable, the user data.

[0072] Also, for example, the generation unit 132 generates a model that outputs a matching score by fixing two variables, the object group and the user data, and varying one variable, the additional information. This matching score indicates the degree of harmony of the combination of the object group, the user data, and the additional information, and also indicates the degree of harmony of the combination of the object group and the user data for each additional information, since the one variable, the additional information, is varied.

[0073] The generation unit 132 generates a model that calculates a matching score by fixing one variable, for example. For example, the generation unit 132 generates a model that outputs a matching score by fixing one variable of the target group and varying two variables, user data and additional information. This matching score indicates the degree of harmony of the combination of the target group, user data, and additional information, and also indicates the degree of harmony of each combination of user data and additional information with the target group by varying the two variables, user data and additional information.

[0074] Also, for example, the generation unit 132 generates a model that outputs a matching score by fixing one variable of the user data and varying two variables of the object group and the additional information. This matching score indicates the degree of harmony of the combination of the object group, the user data, and the additional information, and also indicates the degree of harmony with the combination of the user data for each combination of the object group and the additional information by varying the two variables of the object group and the additional information.

[0075] Also, for example, the generation unit 132 generates a model that outputs a matching score by fixing one variable of the additional information and varying two variables of the target group and the user data. This matching score indicates the degree of harmony of the combination of the target group, the user data, and the additional information, and also indicates the degree of harmony with the combination of the additional information for each combination of the target group and the user data by varying the two variables of the target group and the user data.

[0076] The generation unit 132 generates a model that outputs a matching score by moving all three variables, for example, the target group, the user data, and the additional information. This matching score indicates the degree of harmony of each combination of the target group, the user data, and the additional information.

[0077] (Learning Section 133) The learning unit 133 trains the model generated by the generation unit 132. Specifically, the learning unit 133 trains the model such that the higher the degree of harmony between the combination of the target group, the user data, and the additional information, the higher the matching score.

[0078] The learning unit 133 trains the model so that the closer the input combination of the target group, user data, and additional information is to a combination of a predetermined target group, predetermined user data, and predetermined additional information, the higher the matching score. This predetermined target group is, for example, a target group of a predetermined coordinated image. The learning unit 133 may also train the model taking into consideration rankings. The learning unit 133 may train the model so that the closer the combination is to a combination of a predetermined target group highly rated for a predetermined coordinated image (or a predetermined coordination), predetermined user data related to the predetermined coordinated image, and predetermined additional information related to the predetermined coordinated image, the higher the matching score. Furthermore, the learning unit 133 may train a model so that the matching score is higher on three axes: the poster's coordination image, the poster's (or wearer's) user data, and the annotator data, the closer the combination is to the poster's coordination image, the poster's user data, and the annotator data of an evaluator who gave a high evaluation to the combination of the coordination image and the poster (for example, the user data of the poster's date if the date was successful, and the evaluation data of how many points out of 100 the dater gave it).

[0079] (Calculation unit 134) The calculation unit 134 calculates a matching score using the model generated by the generation unit 132. Specifically, the calculation unit 134 calculates a matching score for each combination by inputting a combination of an object group, user data, and additional information into the model. For example, the calculation unit 134 calculates a matching score for each combination by inputting a combination of an object group selected sequentially from among object group candidates, user data selected sequentially from among user data candidates, and additional information selected sequentially from among additional information candidates into the model.

[0080] (Provider 135) The providing unit 135 provides information on a combination of a target group, user data, and additional information that satisfies a predetermined condition based on the matching score calculated by the calculation unit 134. Specifically, the providing unit 135 provides information on a combination of a target group, user data, and additional information that satisfies a predetermined condition by comparing the matching scores calculated for each combination. For example, the providing unit 135 provides information on one or more combinations with a high matching score. Also, for example, the providing unit 135 provides information on one or more combinations selected from combinations with matching scores exceeding a predetermined threshold.

[0081] [5. Information processing flow] Next, the procedure of information processing by the information processing system 1 according to the embodiment will be described with reference to Fig. 8. Fig. 8 is a flowchart showing the procedure of information processing by the information processing system 1 according to the embodiment.

[0082] As shown in FIG. 8, the information processing device 100 acquires a combination of a target group, user data, and additional information (step S201).

[0083] The information processing device 100 inputs the acquired combination into a model that calculates a matching score based on set matching, and calculates a matching score that indicates the degree of harmony of the acquired combination (step S202).

[0084] The information processing device 100 provides information on combinations that satisfy a predetermined condition based on the matching score calculated for each of the acquired combinations (step S203).

[0085] [6. Processing Variations] The information processing system 1 according to the embodiment described above may be implemented in various different forms other than the above embodiment, so other embodiments of the information processing system 1 will be described below.

[0086] (About the outfit) In the above embodiment, the outfit (corresponding to the "target group" according to the above embodiment) may be information on a combination of items that constitute a coordinated outfit on their own.

[0087] (About the sub-code) In the above embodiment, the sub-coordinate (corresponding to the "target group" according to the above embodiment) may be information on a combination of items that does not constitute a coordinated outfit by itself.

[0088] (About makeup) In the above embodiment, the makeup (corresponding to the "user data" in the above embodiment) may be information on a combination of treatment steps that constitute makeup on their own (for example, a combination of eyebrows, eye shadow, eyeliner, mascara, blush, lipstick, etc.).

[0089] (About sub-makeup) In the above embodiment, the sub-makeup (corresponding to the "user data" according to the above embodiment) may be information on a combination of treatment steps that does not constitute makeup by itself.

[0090] (About the body) In the above embodiment, the body (corresponding to the "user data" according to the above embodiment) may be a combination of information related to body information such as body type, personal color, and face type.

[0091] (About TPO) In the above embodiment, the TPO (corresponding to the "additional information" in the above embodiment) may be information on a combination of when, where, and for what purpose.

[0092] (About advertising) In the above embodiment, the advertisement (corresponding to the "advertising data" according to the above embodiment) may be a combination of advertisements provided to the service user on an SNS or EC site. In this case, multiple advertisements may be displayed on the same screen.

[0093] (for a given combination of data) In the above embodiment, the given combination data may be combination information relating to the service user that the service provider can collect.

[0094] (Regarding the combination data to be searched) In the above embodiment, the search target combination data may be information on combinations that a service provider provides to a service user.

[0095] (Overall service image) In the above embodiment, the information processing device 100 may provide information on the search target combination data based on a set of given combination data. For example, the information processing device 100 may input a set of "outfit" and "makeup" posted by the service user on a predetermined posting site, and may suggest an "advertisement" to be displayed on a screen to which the service user transitions after logging in. Also, for example, the information processing device 100 may input a set of "TPO" and "makeup" of the service user, and suggest an "outfit" using items on a predetermined shopping site, etc. Also, for example, the information processing device 100 may input a set of "TPO" and "outfit" of the service user, and suggest "makeup" using items on a predetermined shopping site, etc. Also, for example, the information processing device 100 may input a set of "TPO" and "sub-outfit" of the service user, and suggest a "sub-outfit" using items on a predetermined shopping site, etc. Also, for example, the information processing device 100 may input a set of "TPO" and "sub-makeup" of the service user, and suggest a "sub-makeup" using items on a predetermined shopping site, etc. Also, for example, the information processing device 100 may input a pair of the service user's "body" and "outfit" and suggest "makeup" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "body" and "makeup" and suggest "outfit" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "body" and "sub-outfit" and suggest "sub-outfit" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "body" and "sub-makeup" and suggest "sub-makeup" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "body" and "TPO" and suggest "outfit" using items on a predetermined shopping site or the like.Also, for example, the information processing device 100 may input a pair of the service user's "body" and "TPO" and suggest "makeup" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "makeup" and "sub-outfit" and suggest a "sub-outfit" using items on a predetermined shopping site or the like. Also, for example, the information processing device 100 may input a pair of the service user's "outfit" and "sub-makeup" and suggest a "sub-makeup" using items on a predetermined shopping site or the like.

[0096] In the above embodiment, the information processing device 100 may provide information on a set of given combination data based on the search target combination data. For example, the information processing device 100 may input the "TPO" of the service user and suggest a set of "outfit" and "makeup" using items on a predetermined shopping site, etc. Also, for example, the information processing device 100 may input the "body" of the service user and suggest a set of "outfit" and "makeup" using items on a predetermined shopping site, etc.

[0097] (How to determine the combination of information to be provided to service users) In the above embodiment, the information processing device 100 may fix the given combination data in the set matching of the three axes, and perform a full search of the remaining axes for possible combination data to be searched. In this case, if the full search is computationally difficult, the information processing device 100 may set a part of the possible combination data to be searched as the search target. For example, the information processing device 100 may perform a full search by not narrowing down the search range for the three axes. For example, when performing set matching on the three axes of sub-outfit 1, sub-outfit 2, and makeup, if sub-outfit 1 is a purchase history and sub-outfit 2 is a product in the cart, the information processing device 100 suggests makeup. In this case, the amount of sub-outfit 1 and sub-outfit 2 is small, so the load of the full search is small, and for makeup, by narrowing down to "the same series of colors as purchased in the past," a highly real-time suggestion is possible. On the other hand, if the makeup is applied to the face in real time, a suggestion of an outfit is possible. For example, if sub-outfit 1 is a product currently on sale and sub-outfit 2 is all products of a brand registered as favorites, there is a high possibility that the user will not be satisfied with the response unless the search is narrowed down by season, material, etc., since there are a large number of both sub-outfits 1 and 2. In this way, even with the same three axes, the presence or absence of narrowing down varies depending on the type of service. In addition, the information processing device 100 may narrow down the search range for all three axes, for example. For example, the information processing device 100 may narrow down some of the axes, or may narrow down all of the axes and search within the narrowed down range.

[0098] (Regarding two-axis set matching) In the above embodiment, the information processing device 100 may take into consideration the style of dressing, for example, by using set matching between the poster's coordinated image (target group) and the poster's style of dressing (style data). This allows the information processing device 100 to propose a style of dressing that matches the coordination of the user U1, for example (determine style data that is compatible with the target group).

[0099] In the above embodiment, the information processing device 100 may use set matching between the poster's coordinated image and other poster's coordinated images to suggest users who should be followed in a coordination service, for example.

[0100] In the above embodiment, the information processing device 100 may use set matching between the user data of the poster (or seller) and the user data of the viewer (or purchaser) to suggest users who should sell items, for example, on a coordination service, a flea market site, etc. This enables the information processing device 100 to appropriately determine to whom an item requested for sale by multiple users should be sold, for example, on a coordination service, a flea market site, etc.

[0101] 7. Effects As described above, the information processing device 100 according to the embodiment includes the calculation unit 134 and the provision unit 135. The calculation unit 134 calculates a set matching score indicating the degree of harmony between the object group, the user data, and the additional information, based on an object group formed by combining a plurality of objects, user data, and additional information. The provision unit 135 provides information on a combination of the object group, the user data, and the additional information that satisfies a predetermined condition, based on the set matching score calculated by the calculation unit 134.

[0102] As a result, the information processing device 100 according to the embodiment can provide information that is highly harmonious on three axes that combines a plurality of target groups, user data, and additional information.

[0103] In addition, when the calculation unit 134 inputs a combination of a target group, user data, and additional information, it calculates a set matching score by inputting a combination of the target group for which the set matching score is to be calculated, the user data for which the set matching score is to be calculated, and the additional information for which the set matching score is to be calculated into a model that outputs a set matching score.

[0104] As a result, the information processing apparatus 100 according to the embodiment can provide highly harmonious information by using a model that calculates a matching score based on set matching.

[0105] In addition, the calculation unit 134 calculates a set matching score for each target group for which the set matching score is to be calculated, using a model that uses the target group, user data, and additional information as variables, fixes the user data and additional information, and sequentially changes the target group, thereby outputting a set matching score that indicates the degree of harmony between the combination of user data and additional information for each target group.

[0106] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing two variables, the user data and the additional information, and moving one variable of the target group.

[0107] In addition, the calculation unit 134 calculates a set matching score for each user data for which the set matching score is to be calculated, using a model that uses the target group, user data, and additional information as variables, fixes the target group and the additional information, and sequentially changes the user data to output a set matching score that indicates the degree of harmony between the combination of the target group and the additional information for each user data.

[0108] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing two variables, the target group and the additional information, and varying one variable of the user data.

[0109] In addition, the calculation unit 134 calculates a set matching score for each piece of additional information that is the subject of the set matching score calculation, using a model that uses the target group, user data, and additional information as variables, fixes the target group and user data, and sequentially changes the additional information to output a set matching score that indicates the degree of harmony between the combination of the target group and user data for each piece of additional information.

[0110] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing two variables, the target group and the user data, and varying one variable of the additional information.

[0111] In addition, the calculation unit 134 calculates a set matching score for each combination of the target group for which the set matching score is to be calculated and the user data for which the set matching score is to be calculated, using a model that uses the target group, user data, and additional information as variables, fixes the additional information, and sequentially changes the combination of the target group and user data, thereby outputting a set matching score that indicates the degree of harmony with the additional information for each combination of the target group and user data.

[0112] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing the additional information and simultaneously varying the two variables of the target group and the user data.

[0113] In addition, the calculation unit 134 calculates a set matching score for each combination of the target group for which the set matching score is to be calculated and the additional information for which the set matching score is to be calculated, using a model that uses the target group, user data, and additional information as variables, fixes the user data, and sequentially changes the combination of the target group and the additional information, thereby outputting a set matching score that indicates the degree of harmony with the user data for each combination of the target group and the additional information.

[0114] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing user data and simultaneously varying two variables, the target group and the additional information.

[0115] In addition, the calculation unit 134 calculates a set matching score for each combination of the user data for which the set matching score is to be calculated and the additional information for which the set matching score is to be calculated, using a model that uses the target group, user data, and additional information as variables, fixes the target group, and sequentially changes the combination of user data and additional information, thereby outputting a set matching score that indicates the degree of harmony with the target group for each combination of user data and additional information.

[0116] As a result, the information processing device 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by fixing a target group and simultaneously varying two variables, the user data and the additional information.

[0117] In addition, the calculation unit 134 uses the target group, user data, and additional information as variables, and sequentially changes the combination of the target group, user data, and additional information, thereby calculating a set matching score for each combination of the target group, user data, and additional information, using a model that outputs a set matching score for each combination of the target group, user data, and additional information, and calculates a set matching score for each combination of the target group for which the set matching score is to be calculated, the user data for which the set matching score is to be calculated, and the additional information for which the set matching score is to be calculated.

[0118] As a result, the information processing apparatus 100 according to the embodiment can provide information with a high degree of harmony by using a model that calculates a matching score by simultaneously changing three variables, namely, the target group, the user data, and the additional information.

[0119] In addition, the calculation unit 134 calculates the set matching score using a model trained to increase the set matching score the closer the input combination of a target group, user data, and additional information is to a combination of a specified target group, specified user data, and specified additional information.

[0120] As a result, the information processing device 100 according to the embodiment can appropriately perform learning of a model that calculates a matching score using three-axis correct answer data that combines a target group, user data, and additional information.

[0121] In addition, the calculation unit 134 calculates the set matching score using a model trained to increase the set matching score as the input combination of the target group, user data, and additional information approaches a combination of a specific target group that has a high rating for the target group, user data related to the specific target group, and additional information related to the specific target group.

[0122] This allows the information processing device 100 according to the embodiment to appropriately perform learning of a model that calculates a matching score using three-axis correct answer data that takes ranking into consideration.

[0123] In addition, the providing unit 135 provides information on one or more combinations with high set matching scores.

[0124] As a result, the information processing device 100 according to the embodiment can appropriately provide information with a high degree of harmony.

[0125] In addition, the providing unit 135 provides information on one or more combinations selected from among the combinations whose set matching scores exceed a predetermined threshold.

[0126] As a result, the information processing apparatus 100 according to the embodiment can appropriately provide information with a high degree of harmony by using a predetermined threshold value.

[0127] The additional information is the user data of the user.

[0128] As a result, the information processing apparatus 100 according to the embodiment can provide highly harmonious information using data related to other contributors as additional information.

[0129] The additional information is user data of a user other than the user.

[0130] In this way, the information processing apparatus 100 according to the embodiment can provide highly harmonious information using the evaluation-related data as additional information.

[0131] The additional information is data relating to advertisements.

[0132] As a result, the information processing device 100 according to the embodiment can provide highly harmonious information using the advertisement-related data as additional information.

[0133] The additional information is data regarding the condition of the subject.

[0134] As a result, the information processing device 100 according to the embodiment can provide highly harmonious information using data on the state of the target as additional information.

[0135] [8. Hardware Configuration] The information display device 10 and the information processing device 100 according to the above-described embodiment are realized, for example, by a computer 1000 having a configuration as shown in Fig. 9. Fig. 9 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information display device 10 and the information processing device 100. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, a HDD 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

[0136] The CPU 1100 operates and controls each unit based on a program stored in the ROM 1300 or the HDD 1400. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, programs that depend on the hardware of the computer 1000, and the like.

[0137] The HDD 1400 stores programs executed by the CPU 1100, data used by such programs, etc. The communication interface 1500 receives data from other devices via a predetermined communication network and sends it to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the predetermined communication network.

[0138] The CPU 1100 controls output devices such as a display and a printer, and input devices such as a keyboard and a mouse, via the input / output interface 1600. The CPU 1100 acquires data from the input devices via the input / output interface 1600. The CPU 1100 also outputs generated data to the output devices via the input / output interface 1600.

[0139] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0140] For example, when the computer 1000 functions as the information display device 10 and the information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control units 14 and 130. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from other devices via a predetermined communication network.

[0141] [9. Others] In addition, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0142] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0143] Furthermore, the above-described embodiments can be appropriately combined as long as the processing contents are not contradictory.

[0144] Although some of the embodiments of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be embodied in other forms that incorporate various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the Disclosure of the Invention section.

[0145] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit", etc. For example, an acquisition section can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0146] 1. Information Processing Systems 10 Information display device 11 Communications Department 12 Input section 13 Output section 14 Control section 100 Information processing device 110 Communications Department 120 Storage section 121 Posted information storage unit 122 Model information storage unit 123 User information storage unit 130 Control section 131 Acquisition Department 132 Generation part 133 Learning Department 134 Calculation section 135 Provision Department 141 Receiving unit 142 Transmitter N Network

Claims

1. a calculation unit that calculates a set matching score indicating a degree of harmony between the object group, the user data, and the additional information based on an object group that is a combination of a plurality of objects, the user data, and the additional information; a providing unit that provides information regarding a combination of the target group, the user data, and the additional information that satisfies a predetermined condition based on the set matching score calculated by the calculating unit; 13. An information processing device comprising:

2. The calculation unit is A model that calculates the set matching score by inputting a combination of the target group, the user data, and the additional information for which the set matching score is to be calculated into a model that outputs the set matching score when the combination of the target group, the user data, and the additional information for which the set matching score is to be calculated is input.

2. The information processing apparatus according to claim 1,

3. The calculation unit is The set matching score is calculated for each of the object groups that are the targets of calculation of the set matching score, using the model that outputs the set matching score indicating the degree of harmony between a combination of the user data and the additional information for each of the object groups by fixing the user data and the additional information and sequentially changing the object groups while using the object group, the user data and the additional information as variables.

3. The information processing apparatus according to claim 2.

4. The calculation unit is The set matching score is calculated for each of the user data to be subjected to the calculation of the set matching score by using the model that outputs the set matching score indicating the degree of harmony between the combination of the target group and the additional information for each of the user data by fixing the target group and the additional information and sequentially changing the user data, while using each of the target group, the user data, and the additional information as variables.

3. The information processing apparatus according to claim 2.

5. The calculation unit is The set matching score is calculated for each of the additional information that is the subject of the calculation of the set matching score, using the model that outputs the set matching score indicating the degree of harmony between the combination of the target group and the user data for each of the additional information by fixing the target group and the user data and sequentially changing the additional information, while using each of the target group, the user data, and the additional information as variables.

3. The information processing apparatus according to claim 2.

6. The calculation unit is The set matching score is calculated for each combination of the object group, the user data, and the additional information, which are variables, by fixing the additional information and sequentially changing the combination of the object group and the user data, using the model that outputs the set matching score indicating the degree of harmony with the additional information for each combination of the object group and the user data, and the set matching score is calculated for each combination of the object group, which is the subject of the calculation of the set matching score, and the user data, which is the subject of the calculation of the set matching score.

3. The information processing apparatus according to claim 2.

7. The calculation unit is The set matching score is calculated for each combination of the object group, which is the subject of the calculation of the set matching score, and the additional information, by using the model that outputs the set matching score indicating the degree of harmony with the user data for each combination of the object group and the additional information by fixing the user data and sequentially changing the combination of the object group and the additional information, with each of the object group, the user data, and the additional information being used as variables.

3. The information processing apparatus according to claim 2.

8. The calculation unit is The set matching score is calculated for each combination of the user data to be calculated for the set matching score and the additional information to be calculated using the model that outputs the set matching score indicating the degree of harmony with the target group for each combination of the user data and the additional information by fixing the target group and sequentially changing the combination of the user data and the additional information with the target group as variables, and 3. The information processing apparatus according to claim 2.

9. The calculation unit is The set matching score is calculated for each combination of the target group, the user data, and the additional information, using the model that outputs the set matching score for each combination of the target group, the user data, and the additional information as variables, and calculating the set matching score for each combination of the target group, the user data, and the additional information, which are the targets for calculating the set matching score.

3. The information processing apparatus according to claim 2.

10. The calculation unit is The set matching score is calculated using the model trained so that the closer the input combination of the target group, the user data, and the additional information is to a combination of a predetermined target group, a predetermined user data, and a predetermined additional information, the higher the set matching score is.

3. The information processing apparatus according to claim 2.

11. The calculation unit is The set matching score is calculated using the model trained so that the set matching score becomes higher as the input combination of the target group, the user data, and the additional information becomes closer to a combination of a specific target group highly rated in relation to the target group, the user data related to the specific target group, and the additional information related to the specific target group.

3. The information processing apparatus according to claim 2.

12. The providing unit is Provide information regarding one or more of the combinations with high set matching scores.

2. The information processing apparatus according to claim 1,

13. The providing unit is providing information regarding one or more of the combinations selected from among the combinations for which the set matching score exceeds a predetermined threshold; 13. The information processing apparatus according to claim 12.

14. The additional information is user data of the user.

2. The information processing apparatus according to claim 1,

15. The additional information is user data of a user different from the user.

2. The information processing apparatus according to claim 1,

16. The additional information is data related to an advertisement.

2. The information processing apparatus according to claim 1,

17. The additional information is data regarding a condition of the subject.

2. The information processing apparatus according to claim 1,

18. 1. A computer-implemented information processing method, comprising: A calculation step of calculating a set matching score indicating a degree of harmony between the object group, the user data, and the additional information based on an object group formed by combining a plurality of objects, the user data, and the additional information; a providing step of providing information on a combination of the target group, the user data, and the additional information that satisfies a predetermined condition based on the set matching score calculated by the calculating step; 13. An information processing method comprising:

19. A calculation step of calculating a set matching score indicating a degree of harmony between the object group, the user data, and the additional information based on an object group combining a plurality of objects, the user data, and the additional information; a provision step of providing information on a combination of the target group, the user data, and the additional information that satisfies a predetermined condition based on the set matching score calculated by the calculation step; An information processing program characterized by causing a computer to execute the above.

Images