Perfumery recommendation system, perfumery recommendation method, and program

Abstract

To provide a technique which can recommend an appropriate perfumery to a user.SOLUTION: A perfumery recommendation system includes: perfumery fragrance information acquisition means which acquires fragrance information of a perfumery based on a measurement value obtained by measuring the perfumery with a fragrance sensor; perfumery fragrance information group storage means in which respective pieces of fragrance information of a plurality of perfumeries acquired by the perfumery fragrance information acquisition means are stored as pieces of information which can be relatively compared; user-provided information acquisition means which acquires user-provided information related to user's selection of perfumeries; recommended article determination means which uses at least the user-provided information and the fragrance information to determine one or more perfumeries to be recommended to the user; and presentation means which presents the recommended perfumeries to the user.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a cosmetic product recommendation system, a cosmetic product recommendation method, and a program. [Background technology]

[0002] Currently, a wide variety of cosmetics, including perfumes and air fresheners, are in widespread use, and many users use these cosmetics. Cosmetics aimed at general consumers usually use names and descriptions that describe the scent, and users use these names and descriptions as clues to select the cosmetics they will use.

[0003] However, the way scents are described often varies depending on the person (the provider of the scent), the social background, and the era, which can lead to problems such as the scent being different from the desired scent when selected solely based on the name or description. Furthermore, for products such as perfume, which are actually perceived as a scent that is created by mixing with a person's body odor, the scent name and description of the scent of the scent itself cannot convey the scent that the user will experience when using the product. For this reason, it has traditionally been the practice to check the scent using a sample at an actual sales point before purchasing the scent.

[0004] Meanwhile, an odor sensor (and an odor evaluation device using the same) has been proposed as a technology that makes it possible to indicate odors using objective indices (for example, Patent Document 1). According to the odor evaluation device described in Patent Document 1, an odor sensor measures the amounts and ratios of various causative substances that are the source of an odor, and by associating the output measurement value with information indicating the odor (name or description), it becomes possible to evaluate and describe odors using objective indices (i.e., in a unified manner). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2023-131155 Summary of the Invention [Problem to be solved by the invention]

[0006] However, even if an odor can be described objectively using an odor sensor (odor evaluation device) such as the one described above, since people perceive and describe odors differently, cosmetic product providers cannot provide cosmetics that match user needs simply by providing such objectified names and descriptions.

[0007] In view of the above-mentioned problems, an object of the present invention is to provide a technology that can recommend appropriate cosmetics to a user. [Means for solving the problem]

[0008] In order to solve the above problems, the present invention employs the following configuration: a means for acquiring odor information about the cosmetic product based on measurements of the cosmetic product using an odor sensor; a cosmetic product odor information group storage means for storing each of the plurality of pieces of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; a user-provided information acquisition means for acquiring user-provided information relating to the cosmetic product selected by the user; Using at least the user-provided information and the scent information of the cosmetic product, A recommended product determination means for determining one or more recommended cosmetics; a presentation means for presenting the recommended cosmetics to the user; It is a cosmetic recommendation system that includes

[0009] In this specification, the term "scent" is used in a sense that is not linked to an evaluation of its quality. Furthermore, "scent information" refers to information that identifies a particular scent, such as the name of the scent, information about the scent's image, such as information indicating the scent type, the names of the substances that cause the scent, and their combinations (ratios). Multiple scent information may be associated with one cosmetic product.

[0010] Another example of "storing each of the scent information of the multiple cosmetics as information that allows for relative comparison" is storing multiple different scent information as mapping information plotted on coordinates in the same dimension.

[0011] The user-provided information may also include odor information related to the user's body odor. The user-provided information may also include odor information related to the user's body odor when the user uses the cosmetic product. The user-provided information may also include at least one of information related to the user's odor preferences and information related to the user's odor needs. The user-provided information may also include information about the user's evaluation of the odor when using the cosmetic product.

[0012] Here, "demand-related information" refers to information based on needs separate from preferences, such as use or usage scenarios. Furthermore, the "scent evaluation" may be an evaluation based on a scent classification axis (sweet, spicy, etc.), an evaluation of likes and dislikes, or an evaluation of the difference between defined scent information and one's own sense. Furthermore, the "scent evaluation when the user uses the cosmetic product" may be an evaluation of the scent of the cosmetic product itself, or an evaluation of body odor when using the cosmetic product.

[0013] A system with this configuration can combine information about the scent of cosmetics (objective information) with the individual user's circumstances regarding cosmetic selection, making it possible to appropriately recommend cosmetics that meet each user's desires or that suit each user's body odor.

[0014] The presentation means may also present information identifying each of the cosmetics to the user along with the scent information of the cosmetics. Examples of "information identifying each of the cosmetics" include the product name, product number, and package image of the cosmetics. This allows the user to confirm the scent information that defines the scent of a particular cosmetic, which is expected to provide more accurate feedback based on this information. In other words, by using such accurate feedback (user-provided information), more appropriate recommended cosmetics can be determined.

[0015] The recommendation system may further include a user information storage means for storing at least user characteristic information, which is information in which the user-provided information is reflected in odor information related to the cosmetics used by the user, and the recommended product determination means may determine the recommended cosmetics using the user characteristic information. Note that the "odor information related to the cosmetics used by the user" includes both "odor information of the cosmetics themselves (odor information of the cosmetics)" used by the user and "odor information related to the user's body odor when using the cosmetics."

[0016] According to this configuration, it is possible to obtain information (user characteristic information) customized for each user based on objective information about the smell. By updating the information each time user-provided information is acquired, it becomes possible to determine recommended cosmetics that better match the user's preferences.

[0017] The present invention can also be understood as a cosmetic product recommendation method as follows: a cosmetic odor information acquisition step of acquiring odor information of the cosmetic based on a measurement value obtained by measuring the cosmetic with an odor sensor; a cosmetic product odor information group storage step of storing the plurality of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; a user-provided information acquisition step of acquiring user-provided information related to the cosmetic product selection by a user; a recommended product determination step of determining one or more cosmetics to recommend to the user using at least the user-provided information and the scent information of the cosmetics; a presentation step of presenting the recommended cosmetics to the user; The cosmetic recommendation method includes:

[0018] The present invention can also be understood as a program for causing a computer to execute each step described in such a method, or as a computer-readable recording medium on which such a program is non-transitoryly recorded.

[0019] The present invention can be achieved by combining the above-described configurations and processes as long as no technical contradiction occurs. [Effects of the Invention]

[0020] According to the present invention, it is possible to provide a technology that can recommend appropriate cosmetics to a user. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a schematic diagram showing the overall configuration of a system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing the hardware configuration of the cosmetic recommendation system according to the embodiment. [Figure 3] FIG. 3 is a block diagram showing the functional configuration of the cosmetic recommendation system according to the embodiment. [Figure 4] FIG. 4 is a part of a flowchart showing the flow of processing performed in the system according to the embodiment. [Figure 5] FIG. 5 is another part of the flowchart showing the flow of processing performed in the system according to the embodiment. [Figure 6]FIG. 6 is a part of a flowchart showing the flow of processing performed in a system according to a modified example. [Figure 7] FIG. 7 is another part of the flowchart showing the flow of processing performed in a system according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0022] <Embodiment> Specific embodiments of the present invention will be described below with reference to the drawings. However, unless otherwise specified, the dimensions, shapes, relative positions, etc. of the components described in the embodiments are not intended to limit the scope of the present invention. Furthermore, not all of the components described in the embodiments are necessarily essential to the present invention.

[0023] (Overall system image) FIG. 1 is a schematic diagram showing the overall configuration of a system according to an embodiment. In this system, a cosmetic recommendation system 10, a user terminal 20, an odor sensor device 30, and an odor information database (DB) 40 are connected via a communication network N.

[0024] The cosmetic recommendation system 10 has a function of selecting cosmetic products (e.g., perfumes) that meet the needs of a user seeking such cosmetic products as recommended cosmetic products and presenting them to the user. This function is realized by exchanging information with a user terminal 20 via a communication network N.

[0025] The user terminal 20 is an information terminal owned by the user, such as a smartphone, tablet, or personal computer. The user uses the cosmetic recommendation system 10 via a dedicated application running on the user terminal 20, a general-purpose mailer, a social networking service (SNS), a web browser, or the like.

[0026] The odor sensor device 30 is equipped with multiple sensor elements with different sensitivities to odor-causing substances, and is a device that identifies odors based on the output values ​​(hereinafter also referred to as the "measurement values" of the odor sensor) of each sensor element, which outputs different electrical signals depending on the odor (actually the amount and ratio of various odor-causing substances in a sample).Technology such as that described in Patent Publication No. 2023-131155 can be adopted.

[0027] The odor information DB 40 is a database in which odor information defining various odors is stored in association with odor sensor measurement values. Here, the "odor information" may be the name of an object that describes an odor (e.g., orange, mint, cypress, lavender, etc.), information indicating the odor family (floral, fruity, oriental, spicy, etc.), etc. In any case, since such odor information is associated with odor sensor measurement values, it can be defined as an objective index.

[0028] Next, the cosmetic recommendation system 10 will be described in detail with reference to Figures 2 and 3. Figure 2 is a schematic diagram showing the hardware configuration of the cosmetic recommendation system 10, and Figure 3 is a functional block diagram of the cosmetic recommendation system 10.

[0029] (Hardware configuration) As shown in FIG. 2, the cosmetic recommendation system 10 is a computer having as its hardware configuration components a processor 91, a memory 92, an input interface (IF) 93, an output IF 94, and a communication IF 95, all of which are interconnected by a connection bus 90.

[0030] The processor 91 is configured by an arithmetic processing device such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), or a DSP (Digital Signal Processor).

[0031] The memory 92 may be a flash memory or a RAM (Random Access Memory), though not shown. The memory 92 includes a main storage device such as a Solid State Drive (SSD), an Erasable Programmable ROM (EPROM), a flash memory, a USB memory, and a Secure Digital (SD) memory card, and an auxiliary storage device such as a Solid State Drive (SSD), an Erasable Programmable ROM (EPROM), a flash memory, a USB memory, and a Secure Digital (SD) memory card. The memory 92 stores information such as programs to be executed by the processor 91, data to be processed by the processor 91, and operation setting information. The processor 91 loads and executes the programs stored in the memory 92, thereby realizing the various functional units shown in FIG. 3.

[0032] The input IF93 includes various input devices such as a keyboard, mouse, camera, and microphone. Examples of the output IF 94 include various output devices such as a display, a speaker, and a printer. A touch panel display can also be used as a configuration that serves both as the input IF 93 and the output IF. The communication IF 95 can have an appropriate configuration depending on the connection method with the communication network N to be connected (wired or wireless, connection standard, etc.).

[0033] Any or all of the functions of the cosmetic recommendation system 10 may be realized by a dedicated hardware device. The cosmetic recommendation system 10 may also be realized by multiple computers that can communicate with each other.

[0034] (Functional configuration) 3, the cosmetic recommendation system 10 includes, as its functional units, a recommendation unit 100, a cosmetic DB 110, a cosmetic scent information DB 120, and a user information DB 130. The recommendation unit 100 also includes a cosmetic scent information acquisition unit 101, a user-provided information acquisition unit 102, a recommended cosmetic product determination unit 103, and a recommended cosmetic product presentation unit 104.

[0035] The fragrance and cosmetic DB 110 stores information about the fragrance and cosmetic products handled by the system. For example, for each fragrance and cosmetic product, at least some of the following information is stored: the fragrance and cosmetic product ID, the name of the fragrance and cosmetic product, the type of fragrance and cosmetic product (perfume, hand cream, aroma oil, etc.), price, scent information, scent sensor measurement values, package image, etc.

[0036] The cosmetic product odor information DB 120 stores the cosmetics handled by the system in a manner that allows for relative comparison of each other's odor information. For example, the information stored in the cosmetic product odor information DB 120 may be map data in which sensor measurement values ​​related to the odor of each cosmetic product are subjected to dimensionality reduction, and the odor information corresponding to each cosmetic product is plotted in the same dimension, such as two or three dimensions. In such a map, the odor information of each cosmetic product does not need to be defined for each "point"; a certain area adjacent to each other can be defined with the same odor information. With such a map, for example, if there are cosmetic products A and B, no matter where on the map each cosmetic product (its odor information) is plotted based on the sensor measurement values, the relative positional relationship between cosmetic products A and B can be compared.

[0037] The creation (or updating) of the map data described above can be performed, for example, as a function provided by the recommendation function unit 100. However, this is not limited to this, and the map data may be created by a system other than the cosmetic recommendation system 10.

[0038] User information DB 130 stores information about users. Specifically, for each user, information such as a user ID, name, contact information, preference information (favorite floral or fruit scent, favorite scent type such as mint, floral, spicy, etc.), demand information (main scenarios for using cosmetics such as business, relaxation, dates, etc.), and attribute information (gender, age, occupation, etc.) is stored. In particular, when the system acquires user-provided information about cosmetics used by a user, user characteristic information, which is information in which the user-provided information is reflected in the scent information of the cosmetics used, is stored in user information DB 130.

[0039] Here, the user characteristic information can be, for example, information obtained by changing the distribution of odor information in the map of the cosmetic odor information DB 120, by reflecting the user's evaluation of the odor information defined in the map of the cosmetic odor information DB 120. For example, if user-provided information is obtained in which a user evaluates the odor of a cosmetic product defined as "fruity" in the cosmetic odor information DB 120 as "floral," the user can determine whether the cosmetic product is "floral." The system generates (user-original) map data that fits within the area defined as "floral." At this time, the relative positions of the scent information on the map may be maintained while shifting the entire map, or the "floral" area on the map may be expanded. The user characteristic information may be created, for example, by the recommendation function unit 100. The recommendation function unit 100 may also associate each piece of user characteristic information created in this way with the target user.

[0040] The cosmetic odor information acquisition unit 101 acquires odor information of a cosmetic product from the measurement values ​​obtained by the odor sensor device 30 and the odor information DB 40, and stores (registers) the acquired information in the cosmetic odor information DB 120. Specifically, odor information corresponding to the measurement values ​​obtained by measuring the odor of a new cosmetic product using the odor sensor device 30 is acquired from the odor information DB 40. Note that "corresponding" here does not mean a one-to-one relationship, but rather refers to all information that satisfies a condition (e.g., the difference between the odor information and the associated sensor measurement value is within a predetermined range). In other words, there may be multiple corresponding odor information, and in such a case, the cosmetic odor information acquisition unit 101 acquires multiple pieces of odor information. Furthermore, if odor information corresponding to a measurement value is not defined in the odor information DB 40, the operator may be prompted to define new odor information. Furthermore, at least the odor information DB 40, the cosmetic DB 110, and the cosmetic odor information DB 120 may each be configured as a relational database.

[0041] The user-provided information acquisition unit 102 acquires user-provided information related to the user's selection of cosmetics, and uses this information to change the contents of the user information DB 130 (including not only updates but also new registrations). Here, the "user-provided information" may include odor information about the user's body odor. Furthermore, this odor information may be odor information about the body odor after using the cosmetics (i.e., a mixture of the odor of the cosmetics and the body odor).

[0042] Incidentally, odor information relating to the user's body odor can be acquired by measuring the user's body odor with the odor sensor device 30 and then using the measured value and the information stored in the odor information DB 40. The user's body odor may be measured directly using a small sensor device 30, or may be measured remotely after receiving a sample from the user (for example, a piece of cloth used to wipe the skin around the neck).

[0043] The user-provided information may also include at least one of information about the user's scent preferences (preference information) and information about the user's needs for cosmetics (demand information). The preference information and demand information can be obtained, for example, by conducting a questionnaire with the user via the user terminal 20 and obtaining the responses. Note that the questionnaire may be required when the user starts using the system (when the user registers).

[0044] Furthermore, as described above, the user-provided information may include user evaluations (feedback) on the scents of cosmetics that have actually been used. Such evaluations can also be obtained by conducting a survey of users. As described above, by updating the user characteristic information based on such evaluations, it becomes possible to make more accurate recommendations.

[0045] The recommended cosmetic product determination unit 103 uses the scent information of the cosmetic products and the user-provided information to determine one or more recommended cosmetic products to recommend to the user. Note that the type of cosmetic product that the recommended cosmetic product determination unit 103 recommends (or can recommend) varies depending on the content of the user-provided information held by the system. For example, the recommended cosmetic product determination unit 103 determines cosmetic products that are likely to meet the user's preferences depending on whether the user-provided information includes the user's body odor data, whether it includes information on evaluations of the scent of the cosmetic product when actually used, and whether it includes preference information and demand information.

[0046] The recommended cosmetic product determination unit 103 can determine recommended cosmetics by reading information from the cosmetic product DB 110, the cosmetic scent information DB 120, and the user information DB 130. In this case, by using the user characteristic information contained in the user information DB 130, it is possible to determine recommended products that effectively match the user's sensibilities.

[0047] The recommended cosmetic product determination unit 103 can be configured to include an estimation model generated by, for example, machine learning. Machine learning algorithms that can be used to create the model include regression analysis, classification, tree, time series analysis, neural networks, and clustering. Examples of regression analysis include logistic regression, Lasso regression, elastic net regression, support vector regression (SVR), linear regression, Ridge regression, and ensemble regression. Examples of classification include k-nearest neighbor (k-nearest neighbor) regression. Examples of the tree include a decision tree, a regression tree, a classification tree, a random forest, boosting (lightGBM, XGboost), and stacking. Examples of the time series include AR, MA, ARIMA, and state space. Examples of the neural network include a multilayer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a residual neural network (ResNet), a transformer, and a graph neural network (GNN). Examples of the clustering include a Gaussian mixture model (GMM), k-means, mini k-means, a variational Gaussian mixture model (VBGMM), and kernel approximation. Furthermore, the estimation model may be further trained by machine learning using the user-provided information acquired by the user-provided information acquisition unit 102 as training data.

[0048] The recommended cosmetic product presentation unit 104 transmits information to the user terminal 20 so that the recommended cosmetic products (at least one of them) determined by the recommended cosmetic product determination unit 103 are displayed on the user terminal 20. There are no particular limitations on how the recommended cosmetic products are displayed on the user terminal 20, but for example, information identifying at least one of the recommended cosmetic products (such as name, package image, or product number) can be displayed together with scent information associated with that cosmetic product. Displaying the information in this manner allows the user to confirm what scent information defines the scent of the recommended cosmetic product, which is expected to provide more accurate feedback when obtaining evaluation information on the use of the recommended cosmetic product.

[0049] (Processing flow in the system) Next, the flow of processing performed in the cosmetic recommendation system 10 will be described with reference to Figures 4 and 5. Figures 4 and 5 are flowcharts showing an example of the flow of processing related to cosmetic recommendation performed in the cosmetic recommendation system 10.

[0050] As shown in FIG. 4, first, the cosmetic recommendation system 10 (cosmetic odor information acquisition unit 101) acquires cosmetic odor information (S101) and registers (maps) the cosmetic odor information in the cosmetic odor information DB 120 (S102). Next, the cosmetic recommendation system 10 (user-provided information acquisition unit 102) acquires user-provided information (S103) and reflects the content of the information in the user information DB 130 (S104). If the user is a first-time user, new data is registered, and if the user is already registered, the registered information is updated. Note that the acquisition of user-provided information in step S102 may be performed at a later stage of processing, as described below.

[0051] After acquiring the fragrance information of the cosmetics and the user-provided information, the cosmetics recommendation system The system 10 then proceeds to a process of determining recommended cosmetic product information at an appropriate timing (S105). Note that "appropriate timing" can be set, for example, when a request is received from the user, when the user browses an EC (electronic commerce) site related to cosmetics, when the user-provided information acquisition unit 102 acquires new user-provided information, etc.

[0052] In step S106, the cosmetic recommendation system 10 (recommended cosmetic product determination unit 103) determines whether or not to use odor information related to the user's body odor (hereinafter also referred to as "body odor data") when determining the recommended cosmetic products (S106). Specifically, the system may make an inquiry by presenting a UI (user interface) screen for the user to confirm and obtain a response, or may make the determination based on whether a request to use body odor data has been registered in advance as user-provided information. In any case, the system can determine whether or not to use body odor data in accordance with the user's wishes.

[0053] If it is determined in step S106 that body odor data will not be used, the recommended cosmetic product determination unit 103 determines one or more cosmetics that suit the user as recommended cosmetics based on preference information, demand information, attribute information, etc. (S107), and proceeds to step S113.

[0054] On the other hand, if it is determined in step S106 that the body odor data is to be used, the recommended cosmetic product determination unit 103 then performs a process of determining whether or not the body odor data is stored (S108). Specifically, it may be configured to determine whether or not the body odor data is registered as part of the user information in the user information DB 130.

[0055] If the recommended cosmetic product determination unit 103 determines in step S108 that body odor data is held, the process proceeds to step S110. On the other hand, if it determines in step S108 that body odor data is not held, the cosmetic product recommendation system 10 (user-provided information acquisition unit 102) acquires body odor data (S109) and proceeds to step S110. Specifically, in step S109, the user is requested to provide body odor data by presenting a UI (user interface) screen requesting the user to provide body odor data or by sending a message, and the body odor data can be acquired. The acquired body odor data may also be classified according to a predetermined type. Note that the method for acquiring a user's body odor data has already been explained, so a repeated explanation will be omitted.

[0056] In step S110, the cosmetic recommendation system 10 (recommended cosmetic determination unit 103) uses the body odor data to determine one or more recommended cosmetics (S110). Specifically, for example, the system can identify (determine) which of the pre-classified body odor types the user's body odor falls into, and then, depending on the body odor type, determine as recommended cosmetics cosmetics that have an odor that counteracts (or masks) the user's body odor, cosmetics that blend with the user's body odor to create an odor that matches the user's preferences and needs, etc. Body odors can be classified into, for example, eight patterns (sweat odor, middle-aged oily odor, aging odor, fatigue odor, bad breath, foot odor, diet odor, and body odor due to dietary habits).

[0057] Next, the cosmetic recommendation system 10 (recommended cosmetic determination unit 103) determines whether there is a request to further narrow down the cosmetic products using the body odor data (S111). Specifically, the recommended cosmetic products determined in step S110 may be presented to the user first, and then a request for narrowing down the cosmetic products may be received from the user, or the system may check whether a request for how to utilize the body odor data has been registered in advance as user-provided information.

[0058] If the cosmetic recommendation system 10 (recommended cosmetic determination unit 103) determines in step S111 that there is no request for narrowing down the search results from the user, the process proceeds to step S113. If it is determined in step S111 that the user has requested narrowing down the search results, the cosmetic recommendation system 10 determines one or more cosmetic products that match the user's narrowing down request as recommended cosmetic products (S112), and proceeds to step S113. The specific details of the user's narrowing down may be obtained from the user, for example, by presenting a UI screen for inputting the narrowing down details in step S111, or may be referenced if the details have been registered in advance as user registration information.

[0059] Then, in step S113, the cosmetic recommendation system 10 (recommended cosmetic presentation unit 104) presents information about the recommended cosmetic determined in any one of steps S107, S110, and S112 to the user terminal 20, for example, by transmitting the information to the user terminal 20 (S113).

[0060] The processes from step S105 to step S113 are repeatedly executed while the cosmetic recommendation system 10 is operating. Furthermore, the processes from step S101 to step S104 are also executed not only when the cosmetic recommendation system 10 is started up, but also whenever a new cosmetic product is registered, a new user is registered, user-provided information is acquired, etc. Specifically, information regarding the user's requests, preferences, etc. acquired in step S106 and step S111 may be stored in the user information DB 130 whenever necessary.

[0061] In the above flow, step S111 may be executed prior to step S110. That is, before determining cosmetics that reflect the body odor data, it may be determined whether the user has requested narrowing down the search.

[0062] According to the above-described cosmetic recommendation system 10, it is possible to recommend appropriate cosmetic products to a user using objective information based on sensor measurements and information provided by the user. Furthermore, by updating the information in the user information DB based on the user-provided information, it becomes possible to recommend more appropriate cosmetic products (those that meet the user's needs) based on information that better reflects the sensibilities of each individual user.

[0063] (Variation) The processing flow of the cosmetic recommendation system 10 in the above embodiment is an example, and the process of recommending cosmetic products to users can be modified in various ways. Below, other processing flows in the cosmetic recommendation system 10 will be described with reference to Figures 6 and 7. Figures 6 and 7 are flowcharts showing an example of the flow of processing related to cosmetic product recommendation performed in a modified example.

[0064] In the description of this modification, the same reference numerals as in the above embodiment are used for the configurations and processes already described in the description of the above embodiment, and repeated description will be omitted. As shown in Fig. 6, in this modification, the same processes as in the above embodiment are executed from step S101 to step S105, but the subsequent processes are different.

[0065] In this modification, after the recommended product determination trigger is turned ON, the cosmetic recommendation system 10 (recommended cosmetic determination unit 103) determines whether or not it has body odor data of the user (S201). Specifically, it may determine whether or not the body odor data is registered as part of the user information in the user information DB 130.

[0066] If it is determined in step S201 that the odor information related to the user's body odor is not held, the recommended cosmetic product determination unit 103 determines one or more cosmetics suitable for the user as recommended cosmetics based on preference information, demand information, attribute information, etc. (S202), and proceeds to step S206.

[0067] On the other hand, if it is determined in step S201 that odor information related to the user's body odor is held, the recommended cosmetic product determination unit 103 then executes a process to determine whether preference information or demand information is held as user-provided information (step S203). If it is determined in step S203 that the user's preference information or demand information is not held, the recommended cosmetic product determination unit 103 determines one or more cosmetics with an odor that neutralizes the user's body odor or cosmetics that mask the user's body odor as recommended cosmetics (S204), and proceeds to step S206.

[0068] If it is determined in step S203 that the user's preference information or demand information is held, the recommended cosmetic product determination unit 103 determines one or more cosmetics whose scent when combined with the user's body odor matches the user's preference or demand as recommended cosmetics (S205), and proceeds to step S206. Note that the scent when combined with the user's body odor may be estimated, for example, by inputting sensor measurement values ​​of the user's body odor and sensor measurement values ​​for each cosmetic product into a pre-trained inference model.

[0069] Then, in step S206, the cosmetic recommendation system 10 (recommended cosmetic presentation unit 104) presents information about the recommended cosmetic determined in any one of steps S202, S204, and S205 to the user terminal 20, for example, by transmitting the information to the user terminal 20 (S206).

[0070] The processes from step S105 to step S206 are repeatedly executed while the cosmetic recommendation system 10 is operating. Furthermore, the processes from step S101 to step S104 are also executed not only when the cosmetic recommendation system 10 is started up, but also whenever a new cosmetic product is registered, a new user is registered, user-provided information is acquired, etc.

[0071] According to the cosmetic recommendation system 10 described above, it is possible to automatically determine recommended cosmetic products based on the information held, and quickly present the recommended cosmetic products to the user, without requiring a step of confirming the user's intention after the trigger is turned on.

[0072] <Other> The above description merely exemplifies the present invention, and the present invention is not limited to the specific embodiments described above. Various modifications and combinations of the present invention are possible within the scope of the technical concept. For example, in the above embodiment, an example was described in which cosmetics are mapped according to scent information in the cosmetic scent information DB 120, but this is not necessarily required, and information may be stored in a data table format.

[0073] Furthermore, although the above embodiment has only one user terminal 20, the cosmetic recommendation system 10 can naturally be linked to multiple user terminals 20. Furthermore, a user terminal 20 is not necessarily required to apply the present invention, and it is also possible to install an information processing terminal as the cosmetic recommendation system 10 in a store or the like, acquire user-provided information via the input IF 93, and present recommended cosmetic products via the output IF 94.

[0074] <Appendix 1> A means for acquiring information about the scent of a cosmetic product (101) for acquiring information about the scent of the cosmetic product based on a measurement value obtained by measuring the scent of the cosmetic product with an odor sensor; a cosmetic product odor information group storage means (120) for storing each of the plurality of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; User-provided information for acquiring user-provided information related to the selection of said cosmetics by the user Acquisition means (102); a recommended product determination means (103) that determines one or more cosmetics to recommend to the user using at least the user-provided information and the fragrance information of the cosmetics; a presentation means (104) for presenting the recommended cosmetics to the user; A cosmetic product recommendation system including:

[0075] <Appendix 2> The user-provided information includes odor information related to the user's body odor. Attachment 1: A cosmetic product recommendation system.

[0076] <Appendix 3> the user-provided information includes odor information relating to the user's body odor when the user uses the cosmetic product; Attachment 2: A cosmetic product recommendation system.

[0077] <Appendix 4> The user-provided information includes at least one of information related to the user's scent preferences and information related to the user's scent needs. 4. A cosmetic recommendation system according to any one of appendices 1 to 3.

[0078] <Appendix 5> the presentation means presents to the user information identifying each of the cosmetics together with scent information of the cosmetics. 5. A cosmetic recommendation system according to any one of appendices 1 to 4.

[0079] <Appendix 6> The user-provided information includes information about an evaluation of an odor when the user uses the cosmetic product. 6. A cosmetic recommendation system according to any one of appendices 1 to 5.

[0080] <Appendix 7> The device further includes a user information storage means (130) for storing at least user characteristic information, which is information in which the user-provided information is reflected in the scent information relating to the cosmetic product used by the user, the recommended product determination means determines the recommended cosmetics using the user characteristic information; 7. A cosmetic recommendation system according to any one of appendices 1 to 6.

[0081] <Appendix 8> a cosmetic odor information acquisition step (S101) of acquiring odor information of the cosmetic based on a measurement value obtained by measuring the cosmetic with an odor sensor; a cosmetic product odor information group storage step (S102) for storing the plurality of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; a user-provided information acquisition step (S103) for acquiring user-provided information related to the cosmetic product selection by the user; a recommended product determination step (S107, S110, S112, S202, S204, S205) of determining one or more cosmetics to recommend to the user using at least the user-provided information and the scent information of the cosmetics; a presentation step (S113, S206) of presenting the recommended cosmetics to the user; A cosmetic product recommendation method comprising:

[0082] <Appendix 9> A program for causing a computer to execute each step described in Appendix 8. [Explanation of symbols]

[0083] 10. Cosmetics recommendation system 20. User terminal 30. Odor sensor device 40. Odor Information Database 90 Bus 91 Processor 92...Memory 93 Input IF 94 Output IF 95 Communication Interface 100 Recommended 110...Cosmetics DB 120···Cosmetics scent information database 130 User Information DB N···Communication Network

Claims

1. a means for acquiring odor information about the cosmetic product based on measurements of the cosmetic product using an odor sensor; a cosmetic product odor information group storage means for storing each of the plurality of pieces of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; a user-provided information acquisition means for acquiring user-provided information relating to the cosmetic product selected by the user; a recommended product determination means for determining one or more cosmetics to recommend to the user using at least the user-provided information and the scent information of the cosmetics; a presentation means for presenting the recommended cosmetics to the user; A cosmetic product recommendation system including:

2. The user-provided information includes odor information related to the user's body odor. The cosmetic recommendation system according to claim 1 .

3. the user-provided information includes odor information relating to the user's body odor when the user uses the cosmetic product; The cosmetic recommendation system according to claim 2 .

4. The user-provided information includes at least one of information related to the user's scent preferences and information related to the user's scent needs. The cosmetic recommendation system according to claim 1 .

5. the presentation means presents to the user information identifying each of the cosmetics together with scent information of the cosmetics. The cosmetic recommendation system according to claim 1 .

6. The user-provided information includes information about an evaluation of an odor when the user uses the cosmetic product. The cosmetic recommendation system according to any one of claims 1 to 5.

7. The device further comprises a user information storage means for storing at least user characteristic information, which is information in which the user-provided information is reflected in scent information relating to the cosmetic product used by the user, the recommended product determination means determines the recommended cosmetics using the user characteristic information; The cosmetic recommendation system according to claim 6.

8. a cosmetic odor information acquisition step of acquiring odor information of the cosmetic based on a measurement value obtained by measuring the cosmetic with an odor sensor; a cosmetic product odor information group storage step of storing the plurality of cosmetic product odor information acquired by the cosmetic product odor information acquisition means as information that allows for relative comparison; a user-provided information acquisition step of acquiring user-provided information related to the cosmetic product selection by a user; a recommended product determination step of determining one or more cosmetics to recommend to the user using at least the user-provided information and the scent information of the cosmetics; a presentation step of presenting the recommended cosmetics to the user; A cosmetic product recommendation method comprising:

9. A program for causing a computer to execute the steps set forth in claim 8.

Images