JP7742784B2 - Information providing device, information providing method, and information providing program - Google Patents

Description

The present invention relates to an information provision device, an information provision method, and an information provision program.

Technologies that support outfit coordination are known in the past. For example, there is a technology that generates outfits based on rules regarding fashion combinations.

Japanese Patent Application Laid-Open No. 2013-235528

However, the above-mentioned conventional technology leaves room for improvement in terms of providing useful information to customers (referred to as "users"). For example, conventional technology makes it difficult to suggest hairstyles that match the clothes a user has purchased.

This application has been made in light of the above, and aims to provide an information provision device, an information provision method, and an information provision program that enable useful information to be provided to users.

In order to solve the above-mentioned problems and achieve the objectives, the information provision device of the present invention is characterized by comprising an acquisition unit that acquires item information related to an accessory selected by a user, a calculation unit that calculates a compatibility level indicating the degree of compatibility between the accessory and a hair style based on the item information, and a generation unit that generates provision information to be provided to the user based on the compatibility level.

The information provision method according to the present invention is executed by an information provision device and includes an acquisition step of acquiring item information related to an accessory selected by a user, a calculation step of calculating a compatibility level indicating the degree of compatibility between the accessory and a hair style based on the item information, and a generation step of generating information to be provided to the user based on the compatibility level.

The information provision program of the present invention is characterized in that it causes a computer to execute the following steps: an acquisition step for acquiring item information related to an accessory selected by a user; a calculation step for calculating a compatibility level indicating the degree of compatibility between the accessory and a hair style based on the item information; and a generation step for generating provision information to be provided to the user based on the compatibility level.

This invention can provide useful information to users.

FIG. 1 is a diagram illustrating an example of the configuration of an information providing system according to an embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of the information providing device according to the embodiment. FIG. 3 is a diagram illustrating an example of an item information storage unit according to the embodiment. FIG. 4 is a diagram illustrating an example of a mode information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of a compatibility information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of a provided information storage unit according to the embodiment. FIG. 7 is a diagram showing a specific example of the information providing process according to the embodiment. FIG. 8 is a flowchart illustrating an example of the flow of the information providing process according to the embodiment. FIG. 9 illustrates an example of a hardware configuration.

Below, detailed explanations will be given of the information provision device, information provision method, and information provision program (hereinafter, "embodiments") according to the present application, with reference to the drawings. Note that the information provision device, information provision method, and information provision program according to the present application are not limited to these embodiments. Furthermore, identical components in the following embodiments will be designated by the same reference numerals, and duplicate explanations will be omitted.

[Embodiment]
The configuration of the information providing system 100 according to the embodiment, the configuration of the information providing device 10, a specific example of the information providing process, and the flow of the information providing process will be described below, and finally the effects of this embodiment will be described.

1. Configuration of information provision system 100
The processing of an information provision system 100 according to this embodiment (referred to as "this system" as appropriate) will be described using Fig. 1. Fig. 1 is a diagram showing an example of the configuration of the information provision system 100 according to this embodiment. Below, the example of the configuration of this system 100, the processing of this system 100, the set matching technology applied in this system 100, and the effects of this system 100 will be described in that order.

(1-1. Example of configuration of information providing system 100)
1 includes an information providing device 10, a user terminal 20, a website server 30, and a behavior information database 40. The information providing device 10, the user terminal 20, the website server 30, and the behavior information database 40 are communicably connected via a predetermined communication network (not shown) via wired or wireless communication. Note that the system 100 may include multiple information providing devices 10, multiple user terminals 20, multiple website servers 30, or multiple behavior information databases 40.

(1-1-1. Information providing device 10)
The information providing device 10 is a device that transmits and receives data between the user terminal 20, the website server 30, and the behavior information database 40, and is realized, for example, by a server device or a cloud system. The example in Fig. 1 shows a case where the information providing device 10 is realized by a server device.

(1-1-2. User terminal 20)
The user terminal 20 is a device (computer) used by a user U to browse web pages, conduct internet shopping on the web, etc. The user terminal 20 accepts operations by the user U. The user terminal 20 may be realized, for example, by a smartphone, a tablet terminal, a notebook PC (Personal Computer), a desktop PC, a mobile phone, a PDA (Personal Digital Assistant), etc. The example in FIG. 1 shows a case where the user terminal 20 is realized by a smartphone.

In the following, user terminal 20 may be referred to as user U. In other words, user U may also be read as user terminal 20. User U may also be used to represent multiple customers.

(1-1-3. Website Server 30)
The website server 30 is a device that transmits and receives data to and from the information providing device 10 and the user terminal 20, and is realized, for example, by a server device or a cloud system. The example in Fig. 1 shows a case where the information providing device 10 is realized by a server device. The website server 30 is a device used by a service provider that provides services such as internet shopping on the web to the user U.

(1-1-4. Behavior Information Database 40)
The behavior information database 40 is a storage device that stores behavior information relating to hair behavior, which will be described later. The behavior information database 40 may store the behavior information as part of the website server 30.

(1-2. Processing of Information Providing System 100)
The following describes steps S1 to S7 as the processing of the information providing system 100. Note that steps S1 to S7 below may be executed in a different order. Also, some of the processing of steps S1 to S7 below may be omitted.

(1-2-1. Processing of step S1)
In the present system 100, first, a user U selects an accessory to purchase or an accessory to consider purchasing via the user terminal 20 and transmits the selection of the accessory to the website server 30 (step S1). Here, the accessory refers to clothing, footwear, accessories, etc. provided to the user U by a business using the website server 30 of the present system 100. For example, the accessory refers to clothing such as tops and bottoms, footwear such as shoes and socks, and accessories such as hats, glasses, gloves, rings, necklaces, earrings, and hair ornaments. The accessory may also be a carrying item such as a bag, backpack, or waist pouch, and is not particularly limited.

(1-2-2. Processing of step S2)
In the present system 100, secondly, the information providing device 10 acquires product information about the clothing item selected by the user U from the website server 30 (step S2). Here, the product information is information about the clothing item, such as, but not limited to, a video image of the clothing item, the product name, the type/category of the product, the price, the available sizes/colors, the manufacturer name, the brand name, related products, etc. Also, in the example of FIG. 1, the information providing device 10 acquires the product information from the website server 30, but it may also acquire the product information from the user terminal 20 or a terminal device/database (not shown).

At this time, the information providing device 10 may acquire user information about the user U from the user terminal 20. Here, the user information refers to information including user attributes such as the user U's gender, age, generation, occupation, annual income, place of residence, marital status, whether or not they have children, videos of the user U, categories of interest to the user U, and behavioral history such as the user U's search history and browsing history on websites, purchase history in online shopping, and location information of the user U. Furthermore, the user information may also include information such as screen information of the user terminal 20 and biometric information of the user U, and is not particularly limited.

(1-2-3. Processing of step S3)
Third, in the present system 100, the information providing device 10 refers to the behavior information database 40 and acquires behavior information (step S3). Here, behavior information is information about hair behavior indicating the overall hairstyle, such as hairstyle and hair color. The behavior information may also be information about the behavior of hair using a wig or toupee, and is not particularly limited. In the example of FIG. 1, the information providing device 10 acquires behavior information from the behavior information database 40, but it may also acquire behavior information from the website server 30 or a terminal device or database (not shown).

(1-2-4. Processing of step S4)
Fourth, in the present system 100, the information providing device 10 calculates a compatibility degree indicating the degree of compatibility between the clothing item selected by the user U and the hair style from the acquired item information and appearance information (step S4). For example, the information providing device 10 uses a machine learning model to calculate the compatibility degree so that it takes a numerical value between 0 and 1 depending on the degree of compatibility between the combination of the clothing item and the hair style. In this case, the information providing device 10 calculates the compatibility degree using a machine learning model such as a deep neural network (DNN) that is trained to output a compatibility degree for each combination of the clothing item and the hair style when the information providing device 10 inputs the item information and appearance information related to the clothing item selected by the user U. Alternatively, the information providing device 10 may calculate the compatibility degree based on rules.

The above-mentioned step S4 will be explained using a specific example. For example, the information providing device 10 can calculate the compatibility level using a machine learning model trained for each user U, a machine learning model trained for each attribute of user U, and a machine learning model trained for each situation of user U, as shown below.

The information providing device 10 can calculate the degree of compatibility using a machine learning model trained for each user U. For example, the information providing device 10 can have a machine learning model learn to match the clothes and hairstyles in images that the user U views on a social networking service (SNS) and has given high ratings, such as a "like." As a result, the information providing device 10 can suggest coordinating hairstyles and salons when the user U selects clothes.

In addition, the information providing device 10 can calculate compatibility using a machine learning model trained for each attribute, including user U's hobbies and preferences. For example, the information providing device 10 generates multiple machine learning models for a user who likes "XX (e.g., celebrity, era, type of drama (action, etc.))" and generates correct answer data for each machine learning model. Specifically, the machine learning model is trained using images that have been "liked" by a subject who likes "XX" or images that match "XX" as correct answer data. As a result, when user U accesses a website, the information providing device 10 can select a machine learning model that matches user U's user attributes and suggest hairstyles that are deemed to have a high degree of compatibility by the selected machine learning model.

In addition, the information providing device 10 can calculate compatibility using a machine learning model trained for each situation of user U. For example, for each corporate recruitment situation, such as new graduates, job changes, and mid-career recruitment, the information providing device 10 generates a machine learning model for each of new graduates, job changes, and mid-career recruitment, using images of candidates who passed the document screening process as correct answer data. Therefore, for example, if user U selects "job changes," the information providing device 10 can suggest a hairstyle that is appropriate for the document screening process when changing jobs.

(1-2-5. Processing of step S5)
Fifth, in the present system 100, the information providing device 10 generates information to be provided to the user U based on the calculated compatibility score, the acquired appearance information, and the user information of the user U (step S5). Here, the provided information refers to information about the hair style identified from the hair style indicated by the appearance information, such as a hairstyle, hair color, or other hair style that matches the clothing item selected by the user U, and information about beauty salons, barber shops, etc. that can provide the hair style. In this case, the information providing device 10 identifies hair styles indicated by the appearance information that have a compatibility score equal to or greater than a predetermined threshold, and generates the provided information. For example, the information providing device 10 identifies an image of a hairstyle that has a compatibility score of "0.7" or higher with the top selected by the user U, and generates the provided information about the image and beauty salons that can provide the hairstyle indicated in the image.

The above-mentioned step S5 will now be explained using a specific example. For example, the information providing device 10 can generate information to be provided based on user information about user U, or based on business information about a beauty salon or the like, as shown below.

The information providing device 10 can generate information to be provided based on user attributes indicated in the user information. For example, the information providing device 10 can identify user U's hair color based on an image of user U and suggest a group of hairstyles with relatively similar hair colors. Furthermore, the information providing device 10 may prioritize suggesting candidate hair styles that are similar to preferred hairstyles registered in advance by user U. Furthermore, even if user U's hair style or preferences are unknown, the information providing device 10 can prioritize suggesting images with hair colors that are relatively similar to the hair color of a product image selected by user U on an internet shopping site. Therefore, the information providing device 10 can suggest highly similar hair styles from among the candidates identified from the style information based on user attributes, etc.

The information providing device 10 can also generate information to be provided based on the behavioral history indicated by the user information. For example, if the information providing device 10 can estimate the belongings of user U based on the behavioral history, such as user U's purchase history, it will suggest hairstyles based on those belongings. Specifically, if user U has a history of purchasing a "skirt," the information providing device 10 will obtain from the website server 30 an image of the "T-shirt" selected by user U and an image of a "skirt" previously purchased, and identify suggested hairstyles by combining the "T-shirt" and "skirt" with various hairstyles.

Furthermore, the information providing device 10 can generate provided information based on business information. For example, the information providing device 10 acquires information about beauty salons and other businesses as business information from the website server 30, as well as information about the reservation sites for those beauty salons and other businesses, and generates suggested information that includes that information. As a result, the information providing device 10 can not only suggest hairstyles that match the clothes purchased by the user U, but also provide information about beauty salons and stylists who can achieve those hairstyles.

(1-2-6. Processing of step S6)
Sixth, in the present system 100, the information providing device 10 transmits the generated information to the user terminal 20 of the user U (step S6). For example, the information providing device 10 transmits to the user terminal 20 of the user U information to be provided, including an image of a hairstyle that has a compatibility rating of 0.7 or higher with the top selected by the user U and information on beauty salons that can provide the hairstyle shown in the image. At this time, the information providing device 10 can also provide the generated information to the user U by transmitting the information to the website server 30.

As described above in steps S1 to S6, the information providing device 10 is a device that can analyze, for example, the hairstyle and hair color of a fashion model wearing the clothing item selected by user U, and provide hairstyle, stylist, and beauty salon information (information about beauty salons and barber shops). That is, when an item of clothing is selected in step S1 described above and the fashion model is wearing that clothing item, the information providing device 10 can provide user U with information about similar hairstyles, stylists who have posted that hairstyle, and the beauty salons to which the stylists belong, from the appearance information database 40. In this case, the information providing device 10 may provide reservation information from the above-mentioned beauty salon's reservation site as the beauty salon information. Furthermore, when user U makes a reservation at the above-mentioned beauty salon and the treatment is actually performed, the information providing device 10 can also charge the beauty salon or stylist a customer referral fee.

(1-2-7. Processing of step S7)
Seventh, in the present system 100, the information providing device 10 learns the calculated compatibility degree (step S7). For example, the information providing device 10 uses a machine learning model to learn by backpropagation or the like so that, when information indicating a good compatibility between a clothing item selected by the user U and a hair style is input, the information providing device 10 outputs the compatibility degree of the hair style with the clothing item as a numerical value approaching "1." Furthermore, when information indicating a bad compatibility between a clothing item selected by the user U and a hair style is input, the information providing device 10 uses a machine learning model to learn by backpropagation or the like so that the information providing device 10 outputs the compatibility degree of the hair style with the clothing item as a numerical value approaching "0."

(1-3. Set matching technology applied in the information provision system 100)
Here, the set matching technology applied in the information provision system 100 described above will be described in detail. A technology called set matching is known as an example of a technology for calculating a score indicating the degree of harmony between sets (hereinafter referred to as a "matching score" as appropriate) based on a set of information indicating feature quantities such as images. Set matching is a technology that uses, for example, deep learning, and estimates that the higher the matching score, the higher the compatibility between the sets. Therefore, set matching can quantitatively evaluate the compatibility between the sets. Calculation of a matching score using set matching is disclosed, for example, in "Exchangeable Deep Neural Networks for Set-to-Set Matching and Learning" by Y. Saito, T. Nakamura, H. Hachiya, and K. Fukumizu. Note that set matching is just an example, and embodiments are not limited to those using this technology.

The information providing device 10 applies a technique for calculating matching scores between sets, such as set matching, to calculate a matching score (compatibility) for each combination of clothing item and hair style. Specifically, the information providing device 10 calculates the compatibility between the hair style and each clothing item selected by the user U. At this time, the information providing device 10 can estimate that the higher the compatibility, the higher the compatibility between the clothing item selected by the user U and the hair style. Therefore, the information providing device 10 identifies the hair style that is calculated to have the highest compatibility for a single clothing item. Furthermore, the information providing device 10 estimates that the identified hair style for a single clothing item is the hair style with the highest compatibility among the hair styles included in the style information. This allows the information providing device 10 to quantitatively determine the compatibility between clothing items and hair styles, rather than relying on the user U's subjective judgment of appearance, etc.

The information providing device 10 also trains the machine learning model as follows. For example, the information providing device 10 identifies images posted on SNS that user U has given favorable reviews. Next, the information providing device 10 identifies the hair style and clothing item in the identified image. At this time, the information providing device 10 may use various identification models or may obtain identification results through crowdsourcing, etc. The information providing device 10 uses the above identification result as correct answer data. Alternatively, the information providing device 10 may use the results of identification from images posted by user U as correct answer data. The information providing device 10 then trains the machine learning model so that it outputs a high score when a hair style and clothing item that are correct answer data are input, and outputs a low score when a hair style and clothing item that are not included in the correct answer data are input. At this time, the information providing device 10 may generate such a machine learning model for each attribute of user U or for each user situation. Furthermore, the information providing device 10 may reflect the degree of compatibility taking into account multiple perspectives by manipulating the output values of multiple models as the output of the machine learning model.

(1-4. Effects of the information provision system 100)
As described above, in the information providing system 100, the information providing device 10 acquires product information regarding clothing items selected by the user U, calculates a compatibility score indicating the degree of compatibility between the clothing items selected by the user U and a hair style based on the product information, and generates information to be provided to the user U based on the compatibility score. The information providing device 10 calculates the compatibility score using a machine learning model that has learned the degree of compatibility between a hair style and clothing items worn by a subject with that hair style. Therefore, the system 100 can estimate a hairstyle that matches the clothing viewed by the user U or the clothing the user U wishes to purchase, and suggest hairstyles and stylists to the user U. Furthermore, the system 100 can estimate hairstyles that go well with accessories, shoes, bags, etc. in addition to the clothing selected by the user U, and suggest hairstyles and stylists to the user U. In other words, the system 100 can provide useful information to the user U on an EC (electronic commerce) site or the like.

In addition, in this system 100, the information providing device 10 further acquires user information about the user U who selected the clothing item, identifies the hair style based on the acquired user information, and generates information to be provided. The information providing device 10 also generates information to be provided that includes business information about businesses that can provide hair styles. As a result, this system 100 can suggest to the user U hairstyles that go well with clothing that is highly rated on social media, etc., or clothing worn by celebrities such as entertainers, and can also make more specific suggestions, such as suggestions about stylists who can achieve those hairstyles. In other words, this system 100 makes it possible to provide more specific and useful information to users U of e-commerce sites, etc.

Furthermore, in the present system 100, the information providing device 10 calculates the compatibility level using a machine learning model that has learned the compatibility level between a hair style and an accessory worn by a subject with that hair style. The information providing device 10 calculates the compatibility level using a machine learning model that has learned the compatibility level from video images including the accessory and the hair style, a machine learning model that has learned the compatibility level for each user U, a machine learning model that has learned the compatibility level for each attribute of user U, or a machine learning model that has learned the compatibility level for each situation of user U. Therefore, the present system 100 generates various machine learning models trained using corrective data collected from social media, television, movies, etc., and can suggest hairstyles to user U with higher accuracy. In other words, the present system 100 enables more specific and useful information to be provided to users U of e-commerce sites, etc., with high accuracy.

2. Configuration of information providing device 10
The configuration of the information providing device 10 according to the embodiment will be described with reference to Fig. 2. Fig. 2 is a block diagram showing an example configuration of the information providing device 10 according to the embodiment. As shown in Fig. 2, the information providing device 10 includes a communication unit 11, a storage unit 12, and a control unit 13. The information providing device 10 may also include an input unit (e.g., a keyboard, a mouse, etc.) that accepts various operations from an administrator of the information providing device 10, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

(2-1. Communication unit 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 communication network via a wired or wireless connection, and transmits and receives information to and from various devices.

(2-2. Storage unit 12)
The storage unit 12 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in Fig. 2, the storage unit 12 according to the embodiment has an item information storage unit 12a, a behavior information storage unit 12b, a compatibility information storage unit 12c, and a provided information storage unit 12d. The storage unit 12 stores various information referenced by the control unit 13 when it operates and various information acquired when the control unit 13 operates.

(2-2-1. Article information storage unit 12a)
The item information storage unit 12a stores various information (item information) related to clothing items selected by the user U. An example of information stored in the item information storage unit 12a will now be described with reference to FIG. 3. FIG. 3 is a diagram illustrating an example of the item information storage unit 12a according to the embodiment. In the example of FIG. 3, the item information storage unit 12a has items such as "item ID" and "item information."

"Item ID" indicates identification information for identifying an item of clothing. "Item information" is information about the item of clothing, such as video images of the item of clothing, item name, item type/category, price, available sizes/colors, manufacturer name, brand name, related items, etc.

That is, Figure 3 shows an example in which the item information for the clothing item identified by the item ID "IID#1" is "Item Information #1," and the item information for the clothing item identified by the item ID "IID#2" is "Item Information #2."

(2-2-2. Behavior information storage unit 12b)
The aspect information storage unit 12b stores various information (aspect information) related to hair aspects. An example of the information stored in the aspect information storage unit 12b will be described below with reference to FIG. 4. FIG. 4 is a diagram showing an example of the aspect information storage unit 12b according to the embodiment. In the example of FIG. 4, the aspect information storage unit 12b has items such as "Aspect ID,""AspectInformation," and "Enterprise Information."

"Appearance ID" indicates identification information for identifying the hair appearance. "Appearance information" is information about the hair appearance, indicating the general hairstyle, such as hair style and hair color, as well as information about hair appearances using wigs, etc. "Business information" is information about the attributes of beauty salons and barber shops that can provide hair appearances, and the hairdressers and barbers (stylists) employed at those beauty salons and barber shops.

In other words, Figure 4 shows an example in which the hair style identified by the style ID "HID#1" has style information "Style Information #1" and business information "Business Information #1," while the hair style identified by the style ID "HID#2" has style information "Style Information #2" and business information "Business Information #2."

(2-2-3. Compatibility Information Storage Unit 12c)
The compatibility information storage unit 12c stores the compatibility (compatibility information) for each hair style with respect to an item of clothing calculated by the calculation unit 13b of the control unit 13. Here, an example of information stored in the compatibility information storage unit 12c will be described with reference to FIG. 5. FIG. 5 is a diagram showing an example of the compatibility information storage unit 12c according to the embodiment. In the example of FIG. 5, the compatibility information storage unit 12c has items such as "item ID,""styleID," and "compatibility."

"Item ID" indicates identification information for identifying an item of clothing. "Style ID" indicates identification information for identifying a hair style. "Compatibility" is a numerical value indicating the degree of compatibility of each hair style with an item of clothing.

In other words, Figure 5 shows an example in which the clothing item identified by the item ID "IID#1" has aspect information "Aspect Information #1" and a compatibility level "Compatibility Level #1," while the clothing item identified by the item ID "IID#2" has aspect information "Aspect Information #2" and a compatibility level "Compatibility Level #2."

(2-2-4. Provided information storage unit 12d)
The provided information storage unit 12d stores the provided information generated by the generation unit 13c of the control unit 13. An example of information stored in the provided information storage unit 12d will now be described with reference to FIG. 6. FIG. 6 is a diagram illustrating an example of the provided information storage unit 12d according to the embodiment. In the example of FIG. 6, the provided information storage unit 12d has fields such as "item ID" and "provided information."

"Item ID" indicates identification information for identifying the clothing item. "Provided information" is information about hair styles, hair colors, and other hair styles that go well with the clothing item selected by user U, and about beauty salons, barber shops, etc. that can provide the hair styles.

In other words, Figure 6 shows an example in which the provided information about the clothing item identified by the item ID "IID#1" is "Provided Information #1," and the provided information about the clothing item identified by the item ID "IID#2" is "Provided Information #2."

(2-3. Control unit 13)
The control unit 13 is realized by, for example, a central processing unit (CPU) or a micro processing unit (MPU) executing various programs (corresponding to an example of an information processing program) stored in a storage device inside the information providing device 10 using RAM as a work area. The control unit 13 is also realized by, for example, an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

As shown in FIG. 2, the control unit 13 has an acquisition unit 13a, a calculation unit 13b, a generation unit 13c, a transmission unit 13d, and a learning unit 13e, and realizes or executes the information processing functions and actions described below. Note that the internal configuration of the control unit 13 is not limited to the configuration shown in FIG. 2, and may be other configurations as long as they perform the information processing described below. Furthermore, the connection relationships between the processing units in the control unit 13 are not limited to the connection relationships shown in FIG. 2, and may be other connection relationships.

(2-3-1. Acquisition unit 13a)
The acquisition unit 13a acquires product information related to the clothing item selected by the user. For example, the acquisition unit 13a acquires, from the website server 30, information related to the item selected by the user U, such as clothing, footwear, or accessories.

The acquisition unit 13a also acquires aspect information related to hair aspects. For example, the acquisition unit 13a acquires, from the aspect information database 40, information on hair aspects indicating general hairstyles such as hairstyle and hair color as aspect information. At this time, the acquisition unit 13a may acquire, from the aspect information database 40, aspect information associated by a service provider on the website server 30. The acquisition unit 13a may also acquire, from the website server 30, aspect information including hair aspects that have been registered in advance as hair aspects that are highly compatible with each clothing item.

The acquisition unit 13a also acquires user information about the user U who selected the clothing item.
For example, the acquisition unit 13a acquires user information from the user terminal 20 of user U, such as user attributes including user U's video images, categories of interest, etc., and user U's behavioral history such as search history, browsing history, purchase history, and location information.

Furthermore, the acquisition unit 13a may acquire business information including information on registered hair salons and stylists from the website server 30. The acquisition unit 13a can also acquire business information including information on hair salons and stylists associated with hair styles from the style information database 40.

The acquisition unit 13a stores the acquired product information in the product information storage unit 12a. The acquisition unit 13a also stores the acquired behavior information and business information in the behavior information storage unit 12b. Furthermore, the acquisition unit 13a can also store the acquired user information in the storage unit 12.

(2-3-2. Calculation unit 13b)
The calculation unit 13b calculates a compatibility degree indicating the degree of compatibility between a combination of a clothing item and a hair style based on the item information. For example, the calculation method will be described. The calculation unit 13b calculates the compatibility degree using a machine learning model that has learned the degree of compatibility between a combination of a hair style and a clothing item worn by a subject having that hair style. As a specific example, the calculation unit 13b calculates the compatibility degree using a machine learning model such as a DNN that has been trained to output a compatibility degree for each hair style with a clothing item when item information and style information selected by the user U are input. Furthermore, the calculation unit 13b may calculate the compatibility degree for each hair style with a clothing item selected by the user U based on a rule.

Furthermore, regarding the calculated numerical values, the calculation unit 13b calculates the compatibility degree so that it takes a numerical value between 0 and 1 depending on the degree of compatibility for each hair type. Furthermore, the calculation unit 13b may calculate the compatibility degree so that it takes a numerical value between 0 and 100% depending on the degree of compatibility for each hair type, and the range and unit of the calculated numerical value are not particularly limited.

To explain this using a specific example, if the compatibility between item I and hairstyle H1 is "0.9," the compatibility between item I and hairstyle H2 is "0.8," and the compatibility between item I and hairstyle H3 is "0.7," the calculation unit 13b will estimate that hairstyle H1, which has the highest compatibility, has a high compatibility with item I. This allows the calculation unit 13b to appropriately optimize the combination of clothing items and hair styles.

Furthermore, regarding the learning data of the machine learning model used by calculation unit 13b, calculation unit 13b calculates the degree of compatibility using a machine learning model that has learned the degree of compatibility from video images including clothing items and hair styles. As a specific example, calculation unit 13b calculates the degree of compatibility using a machine learning model that has learned the degree of compatibility from video images, still images, etc. that simultaneously show clothing items and hair styles collected from websites, SNS, television, movies, etc.

Furthermore, regarding the type of machine learning model used by the calculation unit 13b, the calculation unit 13b calculates the compatibility level using a machine learning model that has learned the degree of compatibility for each user U. As a specific example, the calculation unit 13b calculates the compatibility level using a machine learning model that has learned to match the clothes and hairstyles of images that the user U has viewed on SNS and given high ratings such as "likes."

Furthermore, the calculation unit 13b calculates the compatibility level using a machine learning model that has learned the degree of compatibility for each attribute of the user U. As a specific example, the calculation unit 13b calculates the compatibility level using multiple machine learning models generated for each user attribute, including the user U's age, favorite celebrities, types of dramas, and other hobbies and interests.

The calculation unit 13b also calculates the degree of compatibility using a machine learning model that has learned the degree of compatibility for each situation of the user U. As a specific example, the information providing device 10 calculates the degree of compatibility for each situation, such as new graduate, job change, mid-career recruitment, etc., using a machine learning model that has learned images of candidates who passed document screening during their job hunting activities.

The calculation unit 13b acquires item information related to the clothing item selected by the user U from the item information storage unit 12a. The calculation unit 13b also acquires hair style information indicating the hair style from the style information storage unit 12b. Meanwhile, the calculation unit 13b stores the calculated compatibility degree in the compatibility degree information storage unit 12c.

(2-3-3. Generation unit 13c)
Based on the calculated compatibility, the generation unit 13c generates information to be provided to the user U. For example, the generation unit 13c identifies hair styles having a compatibility level equal to or higher than a predetermined threshold among the hair styles indicated by the acquired hair style information, and generates information to be provided. To explain using a specific example, the generation unit 13c identifies hairstyles H1, H2, H3, ..., which are hair styles for an item I that is a top clothing item, and identifies hairstyles H1, H2, and H3 having a compatibility level equal to or higher than "0.7" with the item I, and generates information to be provided that includes images of the hairstyles H1, H2, and H3.

The generation unit 13c also identifies the hair style based on the acquired user information of the user U and generates information to be provided. For example, the generation unit 13c identifies a hair style registered as a preference of the user U from among hair styles with a compatibility level equal to or greater than a predetermined threshold, and generates information to be provided that includes the hair style.

Furthermore, the generation unit 13c generates provided information including business information about businesses that can provide hair styles. For example, the generation unit 13c generates provided information that includes information about the beauty salon or stylist that provided an image of a hair style with a compatibility level above a predetermined threshold. In this case, the generation unit 13c may generate information about the beauty salon or stylist that includes a link to the reservation page of the beauty salon, etc. In addition, when a user U makes a reservation at the beauty salon, etc., and the treatment is actually performed, the generation unit 13c can also generate information about a customer referral fee to be charged to the beauty salon, etc., or stylist, depending on the number of users U.

The generation unit 13c acquires the compatibility information from the compatibility information storage unit 12c. The generation unit 13c also stores the generated provided information in the provided information storage unit 12d.

(2-3-4. Transmitting unit 13d)
The transmitter 13d transmits the information to be provided generated by the generator 13c to the user U. For example, the transmitter 13d transmits to the user terminal 20 of the user U an image including a hair style whose compatibility with the clothing item selected by the user U is equal to or higher than a predetermined threshold. At this time, the transmitter 13d may transmit, together with the image including the hair style, images of beauty salons, stylists, etc. that can provide the hair style, as well as business information such as available appointment dates.

The transmitting unit 13d obtains the provided information from the provided information storage unit 12d. The transmitting unit 13d may also transmit the provided information to the website server 30 or a business terminal or database (not shown). For example, when a user U makes a reservation at the above-mentioned beauty salon, etc. and the treatment is actually performed, the transmitting unit 13d may transmit information regarding the customer referral fee to be charged to the above-mentioned beauty salon, etc. or stylist.

(2-3-5. Learning unit 13e)
The learning unit 13e learns the machine learning model so as to output the compatibility of each hair style with the clothing item when the product information and style information selected by the user U are input. At this time, the learning unit 13e may learn the machine learning model by backpropagation or the like.

For example, the learning unit 13e trains the machine learning model so that when information indicating a good compatibility between a clothing item selected by user U and a hair style is input to the machine learning model for each user U, user U's attributes, and user U's situation, the compatibility degree of the hair style with the clothing item is output as a numerical value approaching "1."

On the other hand, the learning unit 13e trains the machine learning model so that when information indicating a poor compatibility between the clothing item selected by user U and the hair style is input to the machine learning model for each user U, user U's attributes, and user U's situation, the compatibility degree of the hair style with the clothing item is output as a numerical value approaching "0."

[3. Specific examples of information provision processing]
Next, a specific example of the information provision process according to the embodiment will be described. Hereinafter, the learning process of the machine learning model and the process of generating the information to be provided will be described, and then a specific example of the information to be provided that is transmitted to the user terminal 20 of the user U will be described.

(3-1. Learning process of machine learning model)
The learning process of the machine learning model according to the embodiment will be described. By executing this process, the information providing device 10 can generate a trained machine learning model to be used in the process of generating information to be provided, which will be described later. Below, the supervised data collection process, image data extraction process, and image data learning process will be described in this order.

(3-1-1. Correct data collection process)
First, the information providing device 10 collects correct answer data from SNS. Here, the collected correct answer data is, for example, an image in which a hairstyle and clothing are photographed at the same time, and the evaluation satisfies a predetermined condition. More specifically, the collected correct answer data is an image posted by a predetermined person, an image with more than a predetermined number of "likes," an image with more than a predetermined number of comments or favorable content, etc. Similarly, the information providing device 10 can also collect correct answer data from moving images such as television commercials and movies.

(3-1-2. Image data extraction process)
Next, the information providing device 10 extracts hair and clothing parts (e.g., body parts and belongings) from the collected correct answer data image. At this time, the information providing device 10 may extract clothing parts such as tops, bottoms, accessories, bags, etc.

(3-1-3. Image data learning process)
The information providing device 10 then trains the machine learning model so that when a combination of an image of a hair portion and an image of each clothing portion extracted from the same image (i.e., a combination that is likely to be evaluated well) is input, the machine learning model outputs information indicating high compatibility (e.g., a score of 1). Furthermore, when a combination of an image of a hair portion and an image of a clothing portion extracted from a different image (i.e., a combination that is not known to be evaluated well), the information providing device 10 trains the machine learning model so that the model outputs information indicating low compatibility (e.g., a score of 0 or a random number equal to or less than a predetermined value (0.7)). Through the above-described processing, the information providing device 10 generates a machine learning model that has learned the above-described set matching.

(3-2. Generation process of provided information)
The process of generating information to be provided according to the embodiment will be described. By executing this process, the information providing device 10 can provide hairstyles posted by hairdressers, and when a user U selects a hairstyle, it can provide a reservation service for the beauty salon to which the selected hairdresser belongs (hairdresser reservation service). The following describes the product information acquisition process, candidate information generation process, aspect identification process, and provided information transmission process in this order.

(3-2-1. Product information acquisition process)
First, the information providing device 10 is a server that provides information about, for example, beauty salons. On the other hand, the website server 30 is a server for a website that is, for example, an e-commerce site for clothing and accessories. Here, a user U refers to the website that is an e-commerce site for clothing and accessories. At this time, the website server 30 transmits an image of the product (clothing) that the user U has searched for and selected to the information providing device 10. That is, through the above-described process, the information providing device 10 obtains an image (item information) of the clothing and accessories selected by the user U.

(3-2-2. Candidate Information Generation Process)
Next, the information providing device 10 generates an image by cutting out the product portion from the acquired image, and generates candidate information by combining each hairstyle with the image cut out from the product portion. At this time, the information providing device 10 may accept the hairstyle image (style information) from each hairdresser, or may acquire it from the style information database 40 that collects hairstyle images.

(3-2-3. Mode Identification Processing)
Next, the information providing device 10 inputs each of the generated candidate information into the machine learning model, and identifies a hairstyle included in the candidate information whose output score exceeds a predetermined threshold. That is, through the above-described process, the information providing device 10 identifies a hair style that is highly compatible with the clothing item selected by the user U.

(3-2-4. Provision information transmission process)
Then, the information providing device 10 generates information to be provided by adding a link to a reservation page to the image of the identified hairstyle, and transmits one or more pieces of generated information to the website server 30. As a result, the website server 30 can provide the user U with information such as "Here is a hairstyle that goes well with this outfit. Would you like to make a reservation?" At this time, the information providing device 10 can also transmit the information to be provided directly to the user terminal 20 of the user U.

(3-3. Specific examples of provided information)
A specific example of the information to be provided according to the embodiment will be described with reference to Fig. 7. Fig. 7 is a diagram showing a specific example of the information providing process according to the embodiment. Below, a specific example of the information to be provided that is transmitted to and displayed on the user terminal 20 of the user U will be described.

(3-3-1. Display screen W1)
First, we will explain the display screen W1 related to the clothing or other accessory selected by the user U as the provided information displayed on the user terminal 20 of the user U. The display screen W1 displays an image I1 of a dress (item I), which is the accessory selected by the user U on the EC site, and a screen I2 including the product name, price, available sizes and colors of the accessory, as well as an "Add to Cart" button for transitioning to a purchase screen. At this time, the user U can view information about hairstyles and beauty salons on the display screen W2 (described later) and then decide whether or not to purchase the dress.

(3-3-2. Display screen W2)
Next, we will explain the display screen W2, which contains information displayed on the user terminal 20 of the user U, such as hair styles and other hair styles for clothing and accessories selected by the user U, as well as business information such as beauty salons. The display screen W2 displays hairstyle images H1, H2, and H3 as "hairstyles that go well with this item," along with screens B1, B2, and B3 showing information on beauty salons X, Y, and Z that offer each hairstyle. By clicking on the hairstyle images H1, H2, and H3, the user U can view an enlarged image of each hairstyle or details of the hairstyle. Furthermore, by clicking on screens B1, B2, and B3 showing information on beauty salons X, Y, and Z, the user U can view information on each beauty salon and its associated hairdresser, or transition to a screen for making a reservation at the beauty salon.

[4. Information provision process flow]
The procedure of information processing of the information providing device 10 according to the embodiment will be described with reference to FIG. 8. FIG. 8 is a flowchart showing an example of the flow of the information providing process according to the embodiment. Note that the following steps S101 to S106 may be executed in a different order. Also, some of the following steps S101 to S106 may be omitted.

(4-1. Product information collection process)
First, the acquisition unit 13a of the information providing device 10 executes an item information acquisition process (step S101). For example, the acquisition unit 13a acquires item information related to an item of clothing selected by the user U from the website server 30.

(4-2. Behavior information collection process)
Second, the acquisition unit 13a of the information providing device 10 executes a behavior information acquisition process (step S102). For example, the acquisition unit 13a refers to the behavior information database 40 and acquires behavior information related to the hair behavior.

(4-3. Compatibility Calculation Process)
Third, the calculation unit 13b of the information providing device 10 executes a compatibility calculation process (step S103). For example, the calculation unit 13b calculates the compatibility of each hair style with the clothing item selected by the user U from the acquired item information and style information.

(4-4. Provision information generation process)
Fourth, the generation unit 13c of the information providing device 10 executes a provision information generation process (step S104). For example, the generation unit 13c generates provision information to be provided to the user U based on the calculated compatibility degree and user information.

(4-5. Provision information transmission process)
Fifth, the transmitting unit 13d of the information providing device 10 executes a process of transmitting information to be provided (step S105). For example, the transmitting unit 13d transmits the generated information to be provided to the user terminal 20 of the user U.

(4-6. Compatibility learning process)
Sixth, the learning unit 13e of the information providing device 10 executes a compatibility learning process (step S106). For example, the learning unit 13e learns the calculated compatibility of each hair style with the clothing item selected by the user U.

5. Effects of the embodiment
Finally, the effects of the embodiment will be described below: Effects 1 to 9 corresponding to the processing according to the embodiment will be described below.

(5-1. Effect 1)
In the process according to the embodiment described above, item information about an accessory selected by the user U is acquired, and a compatibility score indicating the degree of compatibility between the accessory and the hair style is calculated based on the acquired item information. Therefore, this process can provide useful information to the user U.

(5-2. Effect 2)
In the process according to the embodiment described above, the compatibility degree is calculated using a machine learning model that has learned the compatibility degree between a hair style and an accessory worn by a subject having the hair style. Therefore, this process can more effectively provide useful information to the user U.

(5-3. Effect 3)
In the process according to the embodiment described above, the degree of compatibility is calculated using a machine learning model that has learned the degree of compatibility from video images including clothing accessories and hair styles. Therefore, this process can provide useful information to the user U more effectively and with higher accuracy.

(5-4. Effect 4)
In the process according to the embodiment described above, the compatibility level is calculated using a machine learning model that has learned the compatibility level for each user U. Therefore, in this process, useful information can be provided to each user U with high accuracy.

(5-5. Effect 5)
In the process according to the embodiment described above, the compatibility degree is calculated using a machine learning model that has learned the compatibility degree for each attribute of the user U. Therefore, in this process, useful information can be provided to the user U with high accuracy for each attribute of the user U.

(5-6. Effect 6)
In the process according to the embodiment described above, the compatibility level is calculated using a machine learning model that has learned the compatibility level for each situation of the user U. Therefore, in this process, useful information can be provided to the user U with high accuracy for each situation of the user U.

(5-7. Effect 7)
In the process according to the present embodiment, the aspect information relating to the hair aspect is further acquired, and among the hair aspects indicated by the aspect information, hair aspects with a compatibility level equal to or higher than a predetermined threshold are identified, and information to be provided is generated. Therefore, by using the aspect information, the process can more effectively provide useful information to the user U.

(5-8. Effect 8)
In the process according to the present embodiment described above, user information about the user U who selected the clothing item is further acquired, the hair style is identified based on the user information, and information to be provided is generated. Therefore, by using the user information, the process can more effectively provide useful information to the user U.

(5-9. Effect 9)
In the process according to the present embodiment described above, provision information including business information about businesses that can provide hair styles is generated, so that useful information including business information can be provided to the user U more effectively.

[Hardware configuration]
The information providing device 10 according to the above-described embodiment is realized by a computer 1000 having a configuration as shown in Fig. 9, for example. The information providing device 10 will be described below as an example. Fig. 9 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to an output device 1010 and an input device 1020, and has a configuration in which a calculation device 1030, a primary storage device 1040, a secondary storage device 1050, an output I/F (Interface) 1060, an input I/F 1070, and a network I/F 1080 are connected via a bus 1090.

The arithmetic unit 1030 operates based on programs stored in the primary storage device 1040 and secondary storage device 1050, programs read from the input device 1020, and the like, and executes various processes. The arithmetic unit 1030 is realized, for example, by a CPU (Central Processing Unit), MPU (Micro Processing Unit), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), etc.

The primary storage device 1040 is a memory device, such as RAM (Random Access Memory), that primarily stores data used by the arithmetic device 1030 for various calculations. The secondary storage device 1050 is a storage device in which data used by the arithmetic device 1030 for various calculations and various databases are registered, and is realized by a ROM (Read Only Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, etc. The secondary storage device 1050 may be internal storage or external storage. The secondary storage device 1050 may also be a removable storage medium such as a USB memory or an SD (Secure Digital) memory card. The secondary storage device 1050 may also be cloud storage (online storage), NAS (Network Attached Storage), a file server, etc.

The output I/F 1060 is an interface for transmitting information to be output to an output device 1010, such as a display, projector, printer, or other device that outputs various types of information, and is implemented, for example, by a connector conforming to a standard such as USB (Universal Serial Bus), DVI (Digital Visual Interface), or HDMI (High Definition Multimedia Interface). The input I/F 1070 is an interface for receiving information from various input devices 1020, such as a mouse, keyboard, keypad, buttons, scanner, or other device, and is implemented, for example, by USB.

Furthermore, the output I/F 1060 and the input I/F 1070 may be wirelessly connected to the output device 1010 and the input device 1020, respectively. In other words, the output device 1010 and the input device 1020 may be wireless devices.

Furthermore, the output device 1010 and the input device 1020 may be integrated, such as a touch panel. In this case, the output I/F 1060 and the input I/F 1070 may also be integrated as an input/output I/F.

The input device 1020 may be a device that reads information from, for example, optical recording media such as CDs (Compact Discs), DVDs (Digital Versatile Discs), and PDs (Phase Change Rewritable Disks), magneto-optical recording media such as MOs (Magneto-Optical Disks), tape media, magnetic recording media, or semiconductor memories.

The network I/F 1080 receives data from other devices via the network N and sends it to the computing device 1030, and also transmits data generated by the computing device 1030 to other devices via the network N.

The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output I/F 1060 and the input I/F 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040 and executes the loaded program.

For example, when the computer 1000 functions as the information providing device 10, the arithmetic unit 1030 of the computer 1000 realizes the functions of the control unit 130 by executing a program loaded onto the primary storage device 1040. The arithmetic unit 1030 of the computer 1000 may also load a program acquired from another device via the network I/F 1080 onto the primary storage device 1040 and execute the loaded program. The arithmetic unit 1030 of the computer 1000 may also link with other devices via the network I/F 1080 and call and use program functions, data, etc. from other programs on the other devices.

〔others〕
Although the embodiments of the present application have been described above, the present invention is not limited to the contents of these embodiments. Furthermore, the above-described components include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the so-called equivalent range. Furthermore, the above-described components can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the spirit of the above-described embodiments.

Furthermore, 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 using known methods. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown.

Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

For example, the information providing device 10 described above may be implemented using multiple server computers, and depending on the function, the configuration can be flexibly changed, such as by calling an external platform using an API (Application Programming Interface) or network computing.

Furthermore, the above-described embodiments and variations can be combined as appropriate to the extent that the processing content is not contradictory.

In addition, the above-mentioned "section, module, unit" can be interpreted as "means" or "circuit." For example, a control unit can be interpreted as a control means or a control circuit.

REFERENCE SIGNS LIST 10 Information providing device 11 Communication unit 12 Storage unit 12a Product information storage unit 12b Behavior information storage unit 12c Compatibility information storage unit 12d Provided information storage unit 13 Control unit 13a Acquisition unit 13b Calculation unit 13c Generation unit 13d Transmission unit 13e Learning unit 20 User terminal 30 Website server 40 Behavior information database 100 Information providing system

Claims (10)

an acquisition unit that acquires item information related to a clothing item selected by a user;
a calculation unit that calculates a compatibility level indicating a degree of compatibility between the clothing item and a hair style based on the item information;
a generation unit that generates information to be provided to the user based on the compatibility degree;
Equipped with
the acquisition unit further acquires preferences of the user as user information related to the user who selected the clothing item;
the generation unit identifies the aspect similar to the user's preference and generates the provided information.
An information providing device characterized by: the calculation unit calculates the compatibility degree using a machine learning model that has learned the compatibility degree between a hair style and a clothing accessory worn by a subject having the hair style;
2. The information providing device according to claim 1. the calculation unit calculates the degree of compatibility using the machine learning model that has learned the degree of compatibility from a video image including clothing accessories and hair styles;
3. The information providing device according to claim 2. the calculation unit calculates the degree of compatibility using the machine learning model that has learned the degree of compatibility for each user;
4. The information providing device according to claim 2 or 3. the calculation unit calculates the compatibility degree using the machine learning model that has learned the compatibility degree for each attribute of the user;
5. The information providing device according to claim 2, wherein the information providing device is a device for providing information to a user. the calculation unit calculates the compatibility degree using the machine learning model that has learned the compatibility degree for each user situation;
6. The information providing device according to claim 2, wherein the information providing device is a device for providing information to a user. The acquisition unit further acquires aspect information relating to a hair aspect,
the generation unit identifies a mode having a compatibility degree equal to or greater than a predetermined threshold among the modes indicated by the mode information, and generates the information to be provided.
7. The information providing device according to claim 1, wherein the information providing device is a device for providing information to a user. the generation unit generates the provision information including business information related to a business that can provide the aspect.
8. The information providing device according to claim 1, wherein the information providing device is a device for providing information to a user. An information providing method executed by an information providing device, comprising:
an acquisition step of acquiring item information relating to a clothing item selected by a user;
a calculation step of calculating a compatibility degree indicating a degree of compatibility between the clothing item and a hair style based on the item information;
a generating step of generating information to be provided to the user based on the compatibility degree;
Including,
the acquiring step further acquires preferences of the user as user information relating to the user who selected the clothing item;
the generating step identifies the aspect similar to the user's preference and generates the provided information.
1. An information providing method comprising: An acquisition step of acquiring item information regarding an apparel item selected by a user;
a calculation step of calculating a compatibility degree indicating a degree of compatibility between the clothing item and a hair style based on the item information;
a generation step of generating information to be provided to the user based on the compatibility degree;
on the computer ,
the obtaining step further includes obtaining a preference of the user as user information relating to the user who selected the clothing item;
the generating step identifies the aspect similar to the user's preference and generates the provided information.
An information providing program characterized by: