KR20230154379A - System and method for supporting user decision-making based on scent of beauty products on metaverse commerce platform - Google Patents

Abstract

In order to support the user's beauty product purchase decision, embodiments verify the user's input preference value based on the user's facial image data when receiving the user's scent information and the user's facial image data. It relates to a system and method that supports user decision-making based on the scent of beauty products in the Metaverse commerce platform.

Description

System and method to support user decision-making based on the scent of beauty products in the Metaverse commerce platform {SYSTEM AND METHOD FOR SUPPORTING USER DECISION-MAKING BASED ON SCENT OF BEAUTY PRODUCTS ON METAVERSE COMMERCE PLATFORM}

Embodiments of this application relate to a system and method for supporting user decision-making based on the scent of beauty products in the Metaverse commerce platform.

Due to the COVID-19 pandemic, wearing masks in stores became mandatory, making it impossible to demonstrate or try out cosmetics, which greatly reduced consumption. In particular, stores and duty-free cosmetics stores in downtown areas such as Myeong-dong and Gangnam, which were booming due to the K-beauty craze, were unable to overcome the high rent burden and closed down as foreign customers stopped visiting.

However, cosmetics are necessities, especially for women, and demand for purchases continues to exist even during the COVID-19 pandemic. Recently, most offline purchases have changed to online, but there are limitations in that traditional non-face-to-face purchases through online shopping malls cannot meet the needs of consumers.

In particular, in the case of beauty products, product performance may be the most important, but it is difficult to immediately feel the product's performance online. Therefore, when choosing a beauty product online, other than experiencing the performance of the product, various purchasing factors such as the beauty product's brand, appearance, color, existing use experience, and recommendations from people around you influence the purchase decision.

Among these, “scent” functions as an important factor in the decision to purchase beauty products. Not to mention perfume, “scent” is very important in purchasing decisions for skin care, makeup, hair care, and body care products.

However, when purchasing a product online, the reality is that it is difficult to even guess about the scent except by checking the appearance or color of the product.

Korean Patent Publication No. 10-2299115 (2021.09.06.)

In order to solve the above-mentioned problems, this application seeks to provide a system and method that supports purchasing decisions considering the scent of beauty products in the Metaverse commerce platform.

A method of supporting user decision-making based on the scent of a beauty product in the Metaverse commerce platform according to one aspect of the present application is performed by one or more computer devices. The method includes obtaining user scent information according to user input in an electronic device, wherein the user's scent information includes the user's scent tendency information and the user's scent preference value including input preference values for scents of a plurality of tasting samples. contains information; generating facial image data of the user by photographing the user's face in response to the scent while inputting the user's input preference value for the scent of each scent sample in the electronic device; When the user's scent information and the user's face image data are received by the service server to support the user's decision to purchase beauty products, the user's input preference value is input based on the user's face image data. It may also include a verification step.

In one embodiment, the verifying step involves verifying the user's input preference value by calculating the user's unconscious preference value for the scent based on the user's face image data, while experiencing each of a plurality of scents. Receiving facial image data of a plurality of captured users; Detecting a face area from the received face image data of a plurality of users; For each of the plurality of scents, measuring biometric information of the user while experiencing each of the plurality of scents from the corresponding detected facial area; The user's biometric information for each of the plurality of scents is input into the emotion recognition model to calculate the user's emotional index for the corresponding scent, and the user's unconscious preference value for the corresponding scent is calculated based on the calculated user's emotional index. steps; And for each of the plurality of scents, it may include verifying the user's input preference value based on the user's unconscious preference value.

In one embodiment, the biometric information may include one or more of heart rate and heart rate variability measured while experiencing each of a plurality of scents. Measuring the user's biometric information includes extracting a G-channel signal from the face area; Converting the G-channel signal to the frequency domain; and measuring one or more heart rate information of heart rate and heart rate variability based on the converted frequency signal.

In one embodiment, the emotion recognition model is a machine learning model designed in advance using the training data set to recognize the input data set and the emotional state of the user corresponding to the input data set, between each training sample and a specific emotional state. As a result of learning the correlation, each training sample in the training set includes biometric information that describes the user's specific emotional state in biometric terms.

In one embodiment, the emotion recognition model is configured to calculate an emotional state value from an input data set and classify the user's emotional state into one of pre-specified classes based on the calculated emotional state value, The emotional state value may represent a probability value that the user of the input data set will be classified into each class.

In one embodiment, the step of verifying the user's input preference value based on the calculated user's unconscious preference value includes calculating the user's first preference ranking for a plurality of scents based on the user's input preference value. step; For each of the plurality of scents, the user's verification preference value is calculated based on the user's input preference value and the user's unconscious preference value for the scent, and the user's second preference for the plurality of fragrances is calculated based on the calculated verification preference value. calculating a ranking; And if the difference between the user's first and second preference rankings for the same scent is smaller than a preset threshold, it may include acknowledging that the user's input preference value has been successfully verified.

In one embodiment, the preferred scent information includes one or more of scent type information, perfume type, and scent level, the user's scent tendency information includes whether or not the user is sensitive to the scent, and the user's preference value for the scent is It is input through a first interface screen, and when the user's scent preference information includes information that the user is sensitive to scent, the first interface screen displays preference for one or more of scent type information, perfume type, and scent level. It is configured to encourage input of a numerical value, and when the user's scent tendency information includes information that the user is not sensitive to scent, the first interface screen displays one or more of scent type information, perfume type, and scent degree. It may be configured to induce input of a preference value.

In one embodiment, the electronic device displays a standby screen including an area displaying the face of the user avatar, and when a user command for selecting a nose portion in the face area of the standby screen is input, the first interface screen is displayed. It may also be configured to display.

In one embodiment, the method includes requesting confirmation of the user's input preference value for a scent for which the verification result failed the user's input preference value; Generating the user's purchase scent history data by mapping the user's preference for scent from the user's preference scent information with the existing beauty product purchase history; To support a purchase decision of a target beauty product having a scent, the target is based on at least a portion of the user's preference scent information including the user's preference values for a plurality of scents reflecting the verification results and the user's purchase scent history data. Calculating a matching score between the scent of a beauty product and the user; And it may further include associating user-customized information with at least some of the scents among the plurality of scents.

In one embodiment, the step of calculating the matching score between the scent of the target beauty product and the user includes retrieving the user's preference value for the scent associated with the target beauty product from the user's preference scent information and the user's purchase scent history data. step; And it may include calculating a matching score between the scent of the target beauty product and the user based on the searched user's preference value. The matching score is calculated from the highest ranking preference value according to the user's preference value reflecting the verification result as the upper limit score.

A computer-readable recording medium according to another aspect of the present application may record a program for performing a method of supporting user decision-making based on the scent of a beauty product in the metaverse commerce platform according to the above-described embodiments.

A system for supporting user decision-making based on the scent of beauty products in the Metaverse commerce platform according to another aspect of the present application includes the user's scent information and the user's scent information to support the user's decision to purchase beauty products. When receiving face image data, it may include a service server that verifies the user's input preference value based on the user's face image data. The user's scent information includes the user's scent preference information and the user's scent preference information including input preference values for the scents of a plurality of tasting samples, and the user's face image data includes the user's scent preference information for the scents of each tasting sample. It was created by photographing the user's face reacting to the scent while entering the input preference value.

The decision support system according to one aspect of the present application collects the user's preference scent information about the scent of the scent sample according to the response to the scent sample delivered to enable domestic and foreign general customers and buyers to purchase beauty products online. It can also support decision making.

In addition, the system maps the user's preference values for scents in the user's preference scent information with the existing beauty product purchase history, databases the user's purchase scent history data, and stores scent information of existing and new beauty products. is linked to one's database, and when purchasing a beauty product, a matching score between the user and the scent can be provided based on some or all of the scent samples, personal purchase scent history data, and scent data of new beauty products.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

In order to more clearly explain the technical solutions of the embodiments of the present invention or the prior art, drawings necessary in the description of the embodiments are briefly introduced below. It should be understood that the drawings below are for illustrative purposes only and not for limiting purposes of the present specification. Additionally, for clarity of explanation, some elements may be shown in the drawings below with various modifications, such as exaggeration or omission.
Figure 1 is a schematic diagram of a system that supports user decision-making based on the scent of a beauty product in the Metaverse commerce platform, according to one aspect of the present application.
Figure 2 is a block diagram of an electronic device according to an embodiment of the present application.
Figure 3 is a block diagram of a service server according to an embodiment of the present application.
Figure 4 is a flowchart of a method for supporting user decision-making based on the scent of a beauty product in the Metaverse commerce platform, according to another aspect of the present application.
Figure 5 shows a tasting sample kit for Allfaction, according to an embodiment of the present application.
Figure 6 is a schematic diagram of a first interface screen for inputting preference values for a trial sample kit into the metaverse commerce platform, which is called from the standby screen, according to an embodiment of the present application.
Figure 7 is a schematic diagram of a process for calculating an unconscious preference value according to an embodiment of the present application.
Figure 8 is a schematic diagram of a process for measuring biometric information according to an embodiment of the present application.
Figure 9 is a network structure diagram of an emotion analysis model according to an embodiment of the present application.
Figure 10 is a schematic diagram of a process for verifying a user's input preference value based on the user's unconscious preference value, according to an embodiment of the present application.
Figure 11 is a schematic diagram of a process for supporting a purchase decision of a target beauty product with a scent using a matching score, according to an embodiment of the present application.

Hereinafter, embodiments of the present application will be examined in detail with reference to the drawings.

However, this disclosure is not intended to limit the disclosure to specific embodiments, and should be understood to include various modifications, equivalents, and/or alternatives to the embodiments of the disclosure. . In connection with the description of the drawings, similar reference numbers may be used for similar components.

In this specification, expressions such as “have,” “may have,” “includes,” or “may include” refer to the corresponding features (e.g., numerical values, functions, operations, steps, parts, elements and/or components). It refers to the presence of components such as etc.) and does not exclude the presence or addition of additional features.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

Expressions such as “first,” “second,” “first,” or “second” used in various embodiments may modify various elements regardless of order and/or importance, and limit the elements. I never do that. The above expressions can be used to distinguish one component from another. For example, the first component and the second component may represent different components, regardless of order or importance.

The expression “configured to” used in this specification may mean, for example, “suitable for,” “having the capacity to,” depending on the situation. ,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured (or set) to” may not necessarily mean “specifically designed to” in terms of hardware. Instead, in some situations, the expression “device configured to” may mean that the device is “capable of” working with other devices or components. For example, the phrase “processor configured (or set) to perform A, B, and C” refers to a processor dedicated to performing those operations (e.g., an embedded processor), or executing one or more software programs stored on a memory device. By doing so, it may mean a general-purpose processor (eg, CPU or application processor) capable of performing the corresponding operations.

In this specification, “metaverse” is a compound word of ‘Meta’, meaning processing and abstraction, and ‘Universe’, meaning the real world, and refers to the existing virtual reality environment. As a more advanced concept than the term, it refers to a virtual space that provides the same or similar experience as the non-real world such as the web and the Internet integrated with the real world. The metaverse may, for example, be implemented as a space that provides augmented or mixed reality to users, but is not limited to this.

Figure 1 is a schematic diagram of a system that supports user decision-making based on the scent of a beauty product in the Metaverse commerce platform, according to one aspect of the present application.

Referring to Figure 1, the system 1 that supports user decision-making based on the scent of beauty products in the metaverse commerce platform includes an electronic device 110, a telecommunication network 120, and a service server 130. .

In the Metaverse commerce platform according to embodiments, the system 1 that supports user decision-making based on the scent of beauty products may be entirely hardware, entirely software, or may have aspects that are partly hardware and partly software. For example, a system may collectively refer to hardware equipped with data processing capabilities and operating software for running it. In this specification, terms such as “unit,” “module,” “device,” or “system” are intended to refer to a combination of hardware and software driven by the hardware. For example, the hardware may be a computing device capable of processing data, including a central processing unit (CPU), a graphics processing unit (GPU), or another processor. Additionally, software may refer to a running process, object, executable, thread of execution, program, etc.

The electronic device 110 is implemented as a terminal device configured to transmit/receive data to and from the service server 130 through the telecommunication network 120. The electronic device 110 includes at least one processor capable of processing data, a memory for storing data, and a communication unit for transmitting/receiving data. The electronic device 110 may be, for example, a laptop computer, notebook, other computing device, tablet, cellular phone, smart phone, smart watch, smart glasses, head mounted display (HMD), other mobile device, or other wearable device. It may be possible.

The telecommunication network 120 provides a wired/wireless telecommunication path through which the electronic device 110 and the service server 130 can transmit and receive data with each other. The telecommunication network 120 is not limited to a communication method according to a specific communication protocol, and an appropriate communication method may be used depending on the implementation. For example, when configured as an Internet Protocol (IP)-based system, the telecommunication network 120 may be implemented as a wired and/or wireless Internet network, and the electronic device 110 and the service server 130 may be used for mobile communication. When implemented as a terminal, the telecommunication network 120 may be implemented as a wireless network such as a cellular network or a wireless local area network (WLAN) network.

The service server 130 is a plurality of computer systems or computer software implemented as network servers. Here, a network server is a computer system and computer that is connected to a sub-device that can communicate with other network servers through a computer network such as a private intranet or the Internet, receives a request to perform a task, performs the task, and provides a performance result. Refers to software (network server program). However, in addition to these network server programs, it should be understood as a broad concept that includes a series of application programs operating on a network server and, in some cases, various databases built within it.

The system 1 is configured to provide a metaverse commerce platform. The Metaverse Commerce Platform provides a Metaverse space for sellers to open stores and performs the function of an e-commerce platform that executes transaction operations between consumers and sellers (e.g., buyers, general customers) through the opened stores. It is implemented to do so. To this end, the system 1 provides the metaverse space of the metaverse commerce platform through the user electronic device 110 and receives user input for transaction operations to be executed on the metaverse space.

Figure 2 is a block diagram of an electronic device according to an embodiment of the present application.

Referring to FIG. 2, the electronic device 110 includes a memory 111, a communication unit 112, a processor 113, a camera 114, an input device 115, and an output device 116.

The memory 111 is connected to the processor 113 and can store data such as basic programs for operation of the processor 113, application programs, setting information, and information generated by operations of the processor 113. The memory 111 may be comprised of volatile memory, non-volatile memory, or a combination of volatile memory and non-volatile memory. Additionally, the memory 111 may provide stored data according to the request of the processor 113. Additionally, the memory 111 may permanently or temporarily store data received from the communication unit 112, data captured by the camera 114, and data generated or processed by the processor 113.

The memory 111 includes HDD (Hard Disk Drive), SSD (Solid State Drive), CD (Compact Disc), RAM (Random Access Memory), Rom (Read Only Memory), and various types of data that store data permanently, semi-permanently, or temporarily. It may also include other storage devices.

In certain embodiments, the memory 111 may store a program (or application) for performing a system or method that supports user decision-making based on the scent of a beauty product in the metaverse commerce platform. The memory 111 provides stored data related to programs for executing the system and method according to instructions of the processor 113.

In the specific embodiments, the memory 111 may store metaverse commerce data received through the communication unit 112, user face image data captured by the camera 114, or computational data generated by the processor 113. It may be possible.

The communication unit 112 is connected to the processor 113 to transmit and receive data, and can transmit and receive data with external devices such as another electronic device 110 or the service server 130. All or part of the communication unit 113 may be referred to as a transmitter, receiver, transceiver, communication unit, communication modem, or communication circuit. The transceiver 112 uses wired access systems and wireless access systems such as the IEEE (institute of electrical and electronics engineers) 802.xx system, IEEE Wi-Fi system, 3rd generation partnership project (3GPP) system, and 3GPP LTE (long term evolution) system. , 3GPP 5GNR (new radio) system, 3GPP2 system, and Bluetooth can support at least one of various wireless communication standards.

The processor 113 may be configured to implement the procedures and/or methods proposed in the present invention. The processor 113 may determine at least one executable operation of the electronic device 110 based on information determined or generated using a data analysis algorithm. And, the processor 113 can control the components of the electronic device 111 to perform the determined operation. To this end, the processor 113 may request, retrieve, receive, or utilize data from the memory 111, and the electronic device 110 may execute a predicted operation or an operation determined to be desirable among the at least one executable operation. ) components can be controlled.

If linkage with an external device such as the service server 130 is necessary to perform the determined operation, the processor 113 generates a control signal to control the external device and sends the signal to the external device through the communication unit 112. The generated control signal can be transmitted.

The processor 113 may be implemented with at least one processor. The processor 113 may include a CPU (Central processing unit) or a neuromorphic processor designed to be advantageous for artificial neural network calculations by incorporating neurons and synapses of the human brain.

The processor 113 is an electronic device 110 to perform transaction operations through the Metaverse Commerce Platform, such as visiting a seller's store, demonstrating a beauty product, and allowing a consumer to purchase a seller's beauty product. ) controls the overall operations.

The camera 114 captures a subject and obtains an image frame, such as a sperm image or video. The acquired image frame may be displayed on the display of the output device 116, transmitted to an external device through the communication unit 112, processed by the processor 113, or stored in the memory 111.

The camera 114 may be comprised of one or more photographing units. The imaging unit may generate an image of the subject in response to wavelengths. The camera 114 may include an RGB sensor to acquire RGB images. Additionally, the camera 114 may further include an IR sensor or a depth sensor to further acquire an IR image or depth image. In certain embodiments, camera 114 may acquire an image that includes the user's face.

The input device 115 is a component configured to receive commands related to input from a user (eg, consumer). The input device 115 may include a touch unit or other input unit.

The touch unit is a component that uses a part of the user's body or another object as a pointing object to input a buyer command. The touch unit may include, but is not limited to, a pressure-sensitive or electrostatic sensor. The other input units include, for example, buttons, keyboards, dials, switches, sticks, etc.

The output device 116 is a component that displays information stored and/or processed in the electronic device 110 and may include, for example, an LCD, OLED, or flexible screen, but is not limited thereto.

Although the input device 115 and the output device 116 are separated in FIG. 2, in many embodiments, the input device 115 and the output device 116 are implemented as a single component to receive input and output information. It can be. For example, the input device 115 and the output device 116 may be a touch panel implemented as a touch screen forming a layer structure with a screen. Touch input is input by a pointing object (eg, including the user's body, a tool, etc.).

In certain embodiments, the electronic device 110 may obtain facial image data of the user through the camera 114. The user's face image includes parts of the user's face. This user's facial image data may be used to calculate the user's perfume preference calculated by the model.

Additionally, the user's electronic device 110 inputs the user's scent preference into the system 1, allowing the user's purchase decision-making activity to proceed on the Metaverse commerce platform.

The operation of this electronic device 110 will be described in more detail with reference to FIGS. 4 to 11 below.

Figure 3 is a schematic block diagram of a service server according to an embodiment of the present application.

Referring to FIG. 3 , the service server 130 is configured to communicate electrically with one or more electronic devices 110 . The service server 130 includes a memory 131, a communication unit 132, and at least one processor 133. Since the components 131, 132, and 133 of FIG. 3 are similar to the components 111, 112, and 113 of FIG. 2, the differences will be mainly described.

The memory 131 may permanently or temporarily store data received from the communication unit 132 and data generated or processed by the processor 133. In certain embodiments, the memory 131 includes metaverse data for implementing the metaverse space provided through the metaverse commerce platform, store data opened in the metaverse space, beauty product data including scent information, and user products. It may also store various information related to the system 1, including purchase history data, user scent preferences, etc.

The processor 133 is configured to perform overall operations for operating the Metaverse commerce platform.

The processor 133 is configured to implement and provide metaverse space and objects such as avatars within the space in two or three dimensions. The processor 133 may use various rendering algorithms to generate metaverse commerce data to implement the metaverse commerce world.

This service server 1 supports purchase decisions for beauty products based on the results of the user's scent preference analysis on the Metaverse commerce platform.

In certain embodiments, the service server 130 may be configured to perform one or more of the operations of claim 1.

Additionally, the service server 130 may include a consultation module that supports determining a user's preferred scent or determining a preference value for the scent. The consultation module is implemented with at least one processor 133 and is configured to execute one or more computer programs that provide conversation content that allows the user to select a preferred scent. Dialogue content is text with context and includes question-type dialogue that elicits user input data (eg, scent preference).

The computer program may be, for example, a natural language program (including, for example, Nature Language Processing (NLP) and/or Nature Language Understanding (NLU) programs) that supports machine learning or artificial intelligence, or various non-humanoid chat programs. . For example, the consultation module may have a chatbot program installed to generate and provide conversation content.

In one embodiment, the counseling module may generate conversation content through a machine learning based chatbot. Machine learning-based chatbots consist of a conversation model that produces output text based on large-scale training samples. The machine learning-based chatbot may be composed of, for example, seq2seq model, pLSA, etc., but is not limited thereto. The training sample may include a guided conversation scenario regarding the user's scent preference information.

In certain embodiments, the service server 130 may include an operation of inputting and managing a user's preferred scent in the avatar setting system of the metaverse commerce platform, an operation of inputting an Allfaction training result based on a scent sample as my scent preference, An operation of generating the user's purchase scent history data by mapping the user's preference for scents in the user's preference scent information with the existing beauty product purchase history. To support purchasing a new beauty product, the user's purchase scent history Based on the data, one or more of the following operations may be performed: calculating a matching score between the scent of a new beauty product and the user, and verifying the user's preference value entered before calculating the matching score through artificial intelligence.

The operation of this service server 130 will be described in more detail with reference to FIGS. 4 to 11 below.

It will be apparent to those skilled in the art that the system 1 may include other components not described herein. For example, it may further include a data input device, an output device such as display and/or printing, a network, a network interface, and a protocol.

Figure 4 is a flowchart of a method for supporting user decision-making based on the scent of a beauty product in the Metaverse commerce platform, according to another aspect of the present application.

The method of FIG. 4 may be performed by one or more computing devices (e.g., system 1).

Referring to FIG. 4 , the method includes acquiring user scent information according to user input in the electronic device 110 (S100). The step (S100) includes obtaining a preference value for the scent of the tasting sample according to the user input.

The user's scent information includes the user's scent tendency information and the user's preferred scent information. Fragrance tendency information includes whether the user is sensitive to scent.

The preferred scent information may include the type and name of the scent that the user normally prefers, and an expression describing the scent, which is unrelated to the target scent of the preference value described below. Expressions describing the scent may include, for example, expressions of olfactory aspects, one's own name or expressions of implicit agreement among market participants (e.g., online nicknames, other keywords, etc.), representative products, etc.

The system 1 provides various interface screens for inputting or managing the user's preferred scent information to the metaverse commerce platform through the electronic device 110.

The electronic device 110 may display a first interface screen for inputting the user's scent information in step S100. The electronic device 110 obtains the user's preferred scent information describing the user's preferred scent through the interface screen. The user's preferred scent information may include the scent name and the user's preference value for the scent. In some embodiments, the user's preference scent information may include the user's preference value for a scent sample set consisting of a plurality of scent samples.

Figure 5 shows a tasting sample kit for Allfaction, according to an embodiment of the present application.

Referring to FIG. 5, the user's preference scent information may include scent preference information for the scent sample. The scent preference information for the tasting sample may include a preference value for the scent of the tasting sample.

Scent preference information for such a tasting sample may be entered based on a user input in response to the scent of the tasting sample for Allfaction delivered to the user. The electronic device 110 may input the user's own preferred scent information according to the results of Allfaction training based on scent samples. As shown in FIG. 5, preference values for scents of a plurality of tasting samples may be input.

In some embodiments, the viewing sample set may include one or more of a first subset, a second subset, and a third subset. The first subset includes one or more tasting samples according to scent type. The second subset includes one or more trial samples according to perfume type. A third subset includes one or more scent samples according to scent level. Then, the preferred scent information may include one or more of scent type information, perfume type, and scent level. The preference value for a scent may also be a preference value for sub-information included in the preference scent information. For example, the preference value includes a preference value for a fragrance type (e.g., fruity, sour, etc.), and/or a preference value for a perfume type (e.g., floral, citrus, etc.). In addition, the preference value includes a preference value for a fragrance. Preference values for degree (e.g., sweety, spicy, etc.) may also be included.

In one example, the set of viewing samples may include some or all of the viewing samples in the following table.

No. A. Scent Type B. Types of perfume C. Fragrance level One flowery floral High 2 fruity citrus Low 3 spicy green powdery 4 Resinous woody Greasy 5 candle (burnt) mint Sweet 6 putrid herb Spicy 7 fruity 8 spicy 9 powdery 10 wateric 11 Sophie 12 balsamic 13 oriental … …

Then, in step S100, the electronic device 110 provides a preference value for at least one item among the plurality of items included in the fragrance type in Table 1, and a preference value for at least one item among the plurality of items included in the perfume type. A preference value for at least one item among a plurality of items included in the numerical value and scent level may be entered through the first interface screen.

Figure 6 is a schematic diagram of a first interface screen for inputting preference values for a trial sample kit into the metaverse commerce platform, which is called from the standby screen, according to an embodiment of the present application.

Referring to FIG. 6, the electronic device 110 may display a standby screen including an area displaying the face of the user avatar. The electronic device 110 may display the first interface screen when a user command to select the nose part in the face area of the standby screen is input.

The first interface screen may include an identification number (eg, #3) of the tasting sample subject to the preference value, a picture visually describing the scent of the tasting sample, and the like.

The preference value may be specified as a user input value in the range of, for example, 0 to 5 or 0 to 10, but the range may be set in various ways.

In step S100, the user may sequentially input his or her preference value for the scent of each tasting sample into the electronic device 110.

In some embodiments, the range of the taste sample of the preference value input through the first interface screen may be based on the user's scent preference information.

If the scent preference information of the first user includes information that the first user is sensitive to scent, the electronic device 110 may be configured to induce input of preference values for all subset categories. For example, the electronic device 110 may be configured to prompt input of preference values for all of the categories of scent type, perfume type, and scent level in Table 1. Through this process, it may be possible to support a sensitive first user to specifically explore his or her preferred scent. In some embodiments, the electronic device 110 may be configured to prompt input of preference values for all subset categories using a consultation module.

If the scent preference information of the second user includes information that the second user is not sensitive to scent, the electronic device 110 may be configured to prompt the user to enter a preference value for some subset of categories in the tasting sample set. It may be possible. For example, the electronic device 110 receives an input for first removing an unfavorable scent from the scent types and perfume types in Table 1, receives an input of preference values for the remaining scents that have not been removed, and/or other It may be configured to perform one or more operations, such as inputting a preference value for a scent included in a list of scents statistically liked by users (eg, top ten scents). One or more of these operations may be performed through the first interface screen.

In this way, different first interface screens may be provided depending on the user's scent preference.

In step S100, the user's scent information, including the user's input preference value for the scent, may be transmitted from the electronic device 110 to the service server 130. The service server 130 uses the user's input preference value for scent to support the user's decision to purchase beauty products.

Referring again to FIG. 4, the method includes generating facial image data of the user in the electronic device 110 while inputting the user's preference value for the scent of each tasting sample in step S100 (S200). , in the service server 130, verifying the user's input preference value input in step S100 based on the user's face image data (S300); It may also include .

The user's face image data generated in step S200 may be transmitted from the electronic device 110 to the service server 130.

In step S300, the service server 110 may calculate the user's unconscious preference value for the scent and verify the user's input preference value for the scent.

The emotions a user feels in response to a scent are expressed through facial expressions. Reactions to scents may vary depending on gender, age, culture, location, time, and season.

Additionally, the emotions felt by the user may differ at least partially depending on the ingredients of the scent. For example, for two scents with different ingredients, a user may feel a high level of happiness, comfort, and excitement for the first scent (e.g., douglas fir) and a high level of happiness, comfort, and excitement for the second scent (e.g., hyacinth). You may experience high levels of happiness, sensuality, and stimulation. The user's reaction through these facial expressions is generally an unconscious reaction in which the user is not aware that “I need to make a specific facial expression for this scent.” The non-conscious preference value based on this non-conscious reaction is a non-input preference value that is not input through the input device 116, as it is a non-conscious result value that the user did not intend to use.

On the other hand, the preference value entered in step S100 is the result of the user's realization that “I should give a certain score to this scent,” and is a self-perceived preference value.

The service server 130 recognizes the user's emotional state and degree of emotional state regarding the scent from the facial expression, and calculates a non-conscious preference value as the user's non-conscious preference reaction to the scent based on the emotion recognition result. You may.

Figure 7 is a schematic diagram of a process for calculating an unconscious preference value according to an embodiment of the present application.

Referring to FIG. 7, the step (S300) of verifying the user's input preference value for a scent by calculating the user's unconscious preference value for the scent is to select each of a plurality of scents (e.g., from the electronic device 110). Receiving facial image data of a plurality of users captured during the experience (S310), detecting a facial area from the received facial image data of a plurality of users (S330), for each of the plurality of scents, the corresponding detected Step of measuring the user's biometric information while experiencing each of the plurality of scents from the face area (S350) Entering the user's biometric information for each of the plurality of scents into an emotion recognition model to calculate the user's emotional index for the corresponding scent and calculating the user's unconscious preference value for the corresponding scent based on the calculated user's emotional index (S370), and for each of the plurality of scents, the user's input preference value based on the user's unconscious preference value. It may also include a verification step (S390).

In step S100, the user sequentially inputs his or her preference values for the scent of each tasting sample into the electronic device 110, while the electronic device 110 responds to the scent of each tasting sample. It is also possible to generate multiple facial image data that captures a user's face with changing expressions. The service server 130 receives a plurality of face image data from the electronic device 110 (S310).

In step 330, the service server 130 may detect the user's face area using various area detection algorithms configured to detect the face area in the input image. In one example, the service server 130 may detect the location of the face area in the input image using deep learning SSD (Single Shot Multibox Detector) technology, but is not limited to this. The service server 130 may detect the user's face area using an area detection algorithm for detecting other face areas.

Region detection algorithms for detecting the face region include, for example, Haar, Convolution Neural Network (CNN), Scale Invariant Feature Transform (SIFT), Histogram of Gradients (HOG), Neural Network (NN), Support Vector It includes, but is not limited to, algorithms that perform Machine (SVM), Gabor method, etc.

In step S350, the user's biometric information may include heart rate and/or heart rate variability. The heart rate and heart rate variability are measured from images instead of sensors while experiencing each of the plurality of scents. The biometric information is used for image-based emotion recognition.

In one embodiment, measuring the user's biometric information (S350) includes extracting a G-channel signal from the face area; Converting the G-channel signal to the frequency domain; And it may include measuring one or more heart rate information of heart rate and heart rate variability based on the converted frequency signal.

In step S330, a G-channel signal is extracted from the patch image representing the detected face area. The G-channel signal may be converted to the frequency domain through FFT processing, and heart rate and/or heart rate variability may be calculated based on the converted signal.

Changes in blood flow are expressed through the G-channel. When converted to the frequency domain, frequencies outside the heart rate range can be filtered out. The service server 130 may measure heart rate and/or heart rate variability using a signal in the heart rate range using a preset standard (eg, NN interval (N-N interval) or RR interval (R-R interval) value).

Figure 8 is a schematic diagram of a process for measuring biometric information according to an embodiment of the present application.

Referring to FIG. 8, when a face area is detected from the user's image data captured during learning by a single RGB camera 114, the detected face area may be composed of pixels of the RGB channel. Pixels representing skin are also expressed as pixels in the RGB channel.

Values for each RGB channel are calculated across all pixels in the face area. When channel-specific values are tracked across all pixels, a 1-D signal for each channel is generated, consisting of channel-specific values and pixels.

The service server 130 extracts each color signal from the face area through R, G, and B channels, and combines each color signal into a single rPPG signal. In some embodiments, the single rPPG signal may be generated using a plane-orthogonal to skin (POS) method. The POS method includes filtering intensity variations by projecting RGB signals on a plane orthogonal to the skin tone vector determined in RGB space; and combining the projection result signals into a one-dimensional signal by adding weights to them. The weight may be determined by the standard deviation ratio of the two signals. This ensures that the rPPG signal in one dimension contains the maximum amount of heart rate components.

In some embodiments, the POS method may further include converting the extracted rPPG signal and facial direction into the frequency domain using Fast Fourier Transform (FFT) and filtering the extracted rPPG signal to the frequency of a human heartbeat.

The 3-axis facial orientation result vector may be converted to a single average value. This average value may be removed from the raw rPPG spectrum after amplitude normalization. In this case, frequency components with high values in the face orientation spectrum are attenuated in the rPPG spectrum, and then frequencies outside the human heartbeat range are removed from the spectrum.

In some embodiments, the POS method may further include a wideband/narrowband filtering step. The strongest frequency component in the spectrum of the resulting signal is used to determine the passband range of the narrow-bandpass filter. The bandwidth of this passband range depends on the peak detection result. The higher the robustness of peak detection, the wider the bandwidth can be. This filter can also be applied to the extracted raw rPPG to generate a noise-free BVP.

Service server 130 may calculate heart rate and/or heart rate variability based on the rPPG signal (raw or filtered). For example, the server 130 may measure heart rate and/or heart rate variability through NN interval (N-N interval) or RR interval (R-R interval) values.

Referring again to FIG. 7, in step S370, the service server 130 may calculate an emotional index from the user's recognition information using an emotion recognition model.

The step (S370) includes, for each of a plurality of scents, calculating the user's emotional index for the scent based on the biometric information using the emotion recognition model, and the user's unconscious preference for the scent. It includes the step of calculating a numerical value.

The emotion recognition model is learned using a training data set consisting of a plurality of training samples. Each training sample in the training set includes biometric information that describes the user's specific emotional state in biometric terms. In some embodiments, each training sample may further include label information indicating the specific emotional state. The training set may include one or more training samples for a specific emotional state.

The emotion analysis model is a machine learning model designed in advance using the training data set to recognize the input data set (e.g., biometric information in step S350) and the emotional state of the user corresponding to the input data set, for each training sample. It is also possible to learn associations between and specific emotional states.

The machine learning model has a neural network structure including fully connected layers. The machine learning model may be composed of, for example, CNN, RNN, or other NN structures, but is not limited thereto.

In some embodiments, the emotion analysis model may be configured to calculate an emotional state value from an input data set and classify the user's emotional state into one of pre-specified classes based on the calculated emotional state value. . The emotional state value may also be referred to as an emotional index. This emotion analysis model is described in more detail with reference to FIG. 9 below.

Figure 9 is a network structure diagram of an emotion analysis model according to an embodiment of the present application.

Referring to FIG. 9, the emotion analysis model may include one or more input layers 510, a plurality of hidden layers 530, and one or more output layers 550. there is.

Each training sample of the training data set is input to the node of the input layer 510. Here, the information 511 included in the training sample corresponds to the basic feature (low level feature) of the emotion analysis model.

At the nodes of the hidden layer 530, calculations are made based on the input parameters. In the plurality of hidden layers 530, the upper hidden layer 530 close to the input layer 510 calculates upper features, and the lower hidden layer 530 close to the output layer 550 calculates lower features. Here, the upper feature and lower feature are features that describe the association between an input data set and a specific emotional state corresponding to the input data set.

The calculation results of these plural hidden layers 530 may be calculated as emotional state values.

The output layer 550 calculates a result of predicting the user's emotional state based on the emotional state value. In some embodiments, the output layer 550 may predict one emotional state among a plurality of pre-designated emotional states as the user's current emotional state.

The plurality of pre-designated emotional states can be classified into positive, normal, or negative emotional categories, and may include, for example, one or more of the following emotional states: joy, surprise, fear, anger, discomfort, sadness, and calmness. The emotional state value may represent the probability of belonging to each class among the seven classes.

In some embodiments, the machine learning model includes an output value (i.e., predicted emotional state) calculated according to each training sample in the training data set, in order to learn the recognition behavior of a specific emotional state corresponding to the input data set. It may also be learned to reduce the difference between the actual values (i.e., the label information of the corresponding sample).

The emotion analysis model learned in this way can also recognize the target user's emotional state in real time.

In step S370, the service server 130 calculates the user's unconscious preference value for the corresponding scent based on the emotional index calculated from the emotion recognition model. The unconscious preference value is calculated as a higher value the higher the positive emotion. also,

The unconscious preference value may be calculated based on a probability value indicating whether the calculated emotional index will be classified into an emotional category such as positive, normal, or negative.

The service server 130 may calculate the user's unconscious preference value through a combination algorithm that calculates a single dependent parameter from a number of independent parameters, such as the sum of probability values for each emotion category and the sum of weights.

In some embodiments, the service server 130 may calculate the user's unconscious preference value for the corresponding scent based on the emotional index for each emotion, which will be classified into each sub-emotion of the emotional category. For example, the service server 130 may calculate the user's unconscious preference value based on probability values for joy, surprise, fear, anger, discomfort, sadness, and calmness.

In some embodiments, the service server 130 converts the calculation value into a range of the user's input preference value (e.g., 0 to 5 in FIG. 6) to facilitate comparison with the input preference value to determine the user's unconscious preference. You can also calculate numbers.

Figure 10 is a schematic diagram of a process for verifying a user's input preference value based on the user's unconscious preference value, according to an embodiment of the present application.

Referring to Figure 10, the step (S390) is a step of calculating the user's first preference ranking for a plurality of scents based on the user's input preference value, and for each of the plurality of scents, the user's input for the scent Calculating a user's verification preference value based on the preference value and the user's unconscious preference value and calculating the user's second preference ranking for a plurality of scents based on the calculated verification preference value, and the user for the same scent If the difference between the first preference ranking and the second preference ranking is smaller than a preset threshold, the step of acknowledging that the user's input preference value has been successfully verified may be included.

Preference can also be interpreted as a quantification of the relative degree of preference. Even if the specific preference values are different, if the non-conscious recognition result and the user input result are ranked as the same or similar preference for a plurality of scents, the user's input preference value may be treated as valid.

The threshold may be a rank difference that can be recognized as equal or similar. In some embodiments, the threshold may be a second-order difference. In Figure 10, the difference between the first preference ranking and the second preference ranking for the fragrances of the tasting sample (#1, #2, #4) has a value of 0 or 1, so the service server 130 selects the scent of the tasting sample. The user's input preference values for (#1, #2, #4) are recognized as valid.

On the other hand, if there is a difference in rankings of 2 or more, the service server 130 does not recognize the user's input preference value. In FIG. 10, the difference between the first preference ranking and the second preference ranking for the scents (#0, #3) of the tasting samples is greater than the threshold value, so the service server 130 selects the scents (#0, #3) of the tasting samples. It may be verified that the user's input preference value for is not valid.

The user's preference values for multiple scents reflecting the verification results are used to support the user's purchase decision.

The user's preference value for the plurality of scents reflecting the verification results may be the user's input preference value or verification preference value.

In some embodiments, the user's input preference value for the scent that successfully verified in step S300 may be used as the user's final preference value for the scent. The service server 130 may support the user's purchase decision using the user's preferred scent information obtained in step S100.

In some other embodiments, the user's verification preference value for the scent that successfully verified in step S300 may be used as the user's final preference value for the scent. The service server 130 may update the user's preference value initially designated as the input preference value from the user's preference scent information obtained in step S100 to the value of the verification preference value. Additionally, the service server 130 may update the user's preference ranking for each scent based on the input preference value to the user's preference ranking based on the verification preference value.

Then, the service server 130 may support the user's purchase decision using the updated verification preference value.

Referring again to FIG. 4, the method may further include a step (S400) of requesting confirmation of the user's input preference value for a scent for which verification failed because the user's input preference value was not valid as a result of the verification. The service server 130 may transmit to the electronic device 110 a request to confirm the user's input preference value for the scents (#0, #3) of the tasting sample that failed verification. The electronic device 110 may display verification failure content and/or confirmation request content and induce the user to re-enter the user's preference value for the scent for which verification failed through the first interface screen (S400).

Additionally, the method may include the step of generating the user's purchase scent history data by mapping the user's preference value for a scent in the user's preference scent information with the existing beauty product purchase history (S500).

In step S500, the service server 130 may map beauty products in the user's transaction history conducted on the Metaverse commerce platform and information on the user's preferred scent.

The user's purchase scent history data may include the user's preference value and preference ranking information for the scent of a beauty product.

In some embodiments, when a plurality of scents are tagged in a mapped beauty product, the service server 130 uses the preference value of each scent tagged in the corresponding beauty product, the highest preference value among the preference rankings, and a representative preference value. , it is also possible to generate user purchase scent history data including rankings.

For example, the purchase scent history data may be generated as shown in the following table.

No. Product Category product main scent my scent data One skin care toner floral, rose Preference Rank 5 2 make up makeup base Powderly, Sophie Preference Rank 3 3 hair care shampoo Citrus, Lemon Preference ranking 7 4 body care body lotion powdery, vanilla, floral Preference Rank 2 5 perfume cologne Citrus, Bergamot, Herbs Preference Rank 6 … … …

The user's purchase scent history data generated in step 500 may be accumulated according to the user's commerce activities. Then, the user's scent preference information included in the user's purchase scent history data is accumulated. For example, scent keywords of scent preference information may be accumulated, such as citrus, woody, floral, rose, cologne, last note, and sophie.

In addition, the method includes at least one of the user's preference scent information and the user's purchase scent history data including the user's preference values for a plurality of scents reflecting the verification results, so as to support a purchase decision of a target beauty product having a scent. It includes a step (S600) of calculating a matching score between the scent of the target beauty product and the user based on the part.

Figure 11 is a schematic diagram of a process for supporting a purchase decision of a target beauty product with a scent using a matching score, according to an embodiment of the present application.

Referring to FIG. 11, the service server 130 may calculate a matching score between the user and the corresponding scent for a target beauty product having a scent and provide the score to the user (S600). The higher the matching score, the more the user prefers the scent of the target beauty product. For example, the user may receive the matching score shown in FIG. 11 from the service server 130 and decide to purchase the lip gloss product of company C with the highest matching score.

In some embodiments, the step (S600) includes retrieving a user's preference value for a scent associated with a target beauty product from the user's preference scent information and the user's purchase scent history data, and based on the retrieved user's preference value. This may include calculating a matching score between the scent of the target beauty product and the user.

In some embodiments, the matching score may be calculated as the upper limit score (eg, 100%) based on the highest ranking preference value according to the user's preference value reflecting the verification result. For example, in an embodiment in which the user's input preference value with successful verification is directly used as the user's preference value, the matching score of the user with five beauty products each having a scent (#0 to #4) is scent (#2) It can also be calculated based on the preference value of “5”. For example, the matching score between a beauty product with scent (#0) and a user is 60% (=3/5), and the matching score between a beauty product with scent (#1) and a user is 40% (=2/5) ), the matching score between the beauty product with scent (#2) and the user is 100% (=5/5), the matching score between the beauty product with scent (#3) and the user is 80% (=4/5), The matching score between the beauty product with scent (#1) and the user may each be calculated as 40% (=2/5).

In some embodiments, when a plurality of fragrances are tagged with a mapped beauty product, the matching score may be calculated based on a representative preference value.

In addition, the method may further include a step (S700) of associating user-customized information with at least some of the scents among the plurality of scents. User-customized information may include, for example, “my name,” “representative product,” etc.

In some embodiments, the scent with which user-customized information can be associated in step S700 may be a scent ranked high in the preference ranking based on the preference value in step S300. The top ranking may be, for example, Top Ten, but is not limited thereto.

For example, the service server 130 may associate “my own name” and “representative product” with a specific type of scent as shown in the table below.

Rank Scent Type my own name representative product One Bergamot, Fig, Cedarwood Saipan summer sea scent Aqua Di Parma, Pico di Amalfi 2 rose Subtle and innocent rose scent Santa Maria Novella Rose Water Toner 3 Monday flower, clover, vanilla, floral The scent of a calm autumn night forest Bulli Body Lotion Mexican Tuberose … … … …

When implementing embodiments of the present invention using hardware, application specific integrated circuits (ASICs) or digital signal processors (DSPs), digital signal processing devices (DSPDs), or programmable PLDs (PLDs) configured to perform the embodiments of the present application. logic devices), field programmable gate arrays (FPGAs), etc. may be included in the components of the present application.

The operation of the system and method for supporting user decision-making based on the scent of beauty products in the metaverse commerce platform according to the embodiments of the present application described above is at least partially implemented as a computer program, and is recorded as a computer-readable record. Can be recorded on media. For example, implemented with a program product comprised of a computer-readable medium containing program code, which can be executed by a processor to perform any or all steps, operations, or processes described.

The computer-readable recording medium includes all types of recording devices that store data that can be read by a computer. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. Additionally, computer-readable recording media may be distributed across computer systems connected to a network, and computer-readable codes may be stored and executed in a distributed manner. Additionally, functional programs, codes, and code segments for implementing this embodiment can be easily understood by those skilled in the art to which this embodiment belongs.

The present invention examined above has been described with reference to the embodiments shown in the drawings, but these are merely illustrative examples, and those skilled in the art will understand that various modifications and modifications of the embodiments are possible therefrom. However, such modifications should be considered within the technical protection scope of the present invention. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached patent claims.

110: electronic device 118: microphone
111: memory 119: sensor
112: Transmitter and receiver 120: Telecommunication network
113: processor 130: server
114: Camera 131: Memory
115: Input device 132: Transmitter and receiver
116: output device 133: processor

Claims (12)

In a method of supporting user decision-making based on the scent of beauty products in the Metaverse commerce platform,
Obtaining user scent information according to user input in an electronic device - the user's scent information includes the user's scent preference information and the user's scent preference information including input preference values for the scents of a plurality of tasting samples. ;
generating facial image data of the user by photographing the user's face in response to the scent while inputting the user's input preference value for the scent of each scent sample in the electronic device;
When the service server receives the user's scent information and the user's face image data to support the user's decision to purchase beauty products, the user's input preference value is verified based on the user's face image data. Including the steps of:
method.
The method of claim 1,
The step of verifying the user's input preference value based on the user's face image data,
Verifying the user's input preference value by calculating the user's unconscious preference value for the scent based on the user's facial image data,
Receiving face image data of a plurality of users captured while experiencing each of the plurality of scents;
Detecting a face area from the received facial image data of a plurality of users;
For each of the plurality of scents, measuring biometric information of the user while experiencing each of the plurality of scents from the detected corresponding facial area;
The user's biometric information for each of the plurality of scents is input into the emotion recognition model to calculate the user's emotional index for the corresponding scent, and the user's unconscious preference value for the corresponding scent is calculated based on the calculated user's emotional index. steps; and
For each of the plurality of scents, comprising the step of verifying the user's input preference value based on the calculated user's unconscious preference value,
method.
According to claim 2,
The biometric information includes one or more of heart rate and heart rate variability measured while experiencing each of the plurality of scents,
The step of measuring the user's biometric information is,
extracting a G-channel signal from the face area;
Converting the G-channel signal to the frequency domain; and
Characterized in that it includes the step of measuring one or more heart rate information of heart rate and heart rate variability based on the converted frequency signal,
method.
According to claim 2,
The emotion recognition model is a machine learning model designed in advance using the training data set to recognize the input data set and the user's emotional state corresponding to the input data set, and learns the correlation between each training sample and a specific emotional state. As,
Each training sample in the training set is characterized in that it includes biometric information that describes the user's specific emotional state in biometric terms.
method.
According to claim 4,
The emotion recognition model is,
configured to calculate an emotional state value from an input data set and classify the user's emotional state into one of pre-specified classes based on the calculated emotional state value;
Characterized in that the emotional state value represents a probability value that the user of the input data set will be classified into each class,
method.
The method of claim 2, wherein the step of verifying the user's input preference value based on the calculated user's unconscious preference value comprises:
calculating the user's first preference ranking for a plurality of scents based on the user's input preference value;
For each of the plurality of scents, the user's verification preference value is calculated based on the user's input preference value and the user's unconscious preference value for the scent, and the user's second preference for the plurality of fragrances is calculated based on the calculated verification preference value. calculating a ranking; and
Characterized in that it includes the step of recognizing the user's input preference value as successful when the difference between the user's first preference rank and the second preference rank for the same scent is smaller than a preset threshold,
method.
The method of claim 1,
The preferred scent information includes one or more of scent type information, perfume type, and scent level,
The user's scent tendency information includes whether or not the user is sensitive to scent,
The user's scent preference value is input through the first interface screen,
When the user's scent preference information includes information that the user is sensitive to scent, the first interface screen is configured to encourage input of preference values for one or more of scent type information, perfume type, and scent level, ,
When the user's scent preference information includes information that the user is not sensitive to scent, the first interface screen prompts the user to enter a preference value for one or more of scent type information, perfume type, and scent level. Characterized in that it consists of,
method.
According to claim 7,
The electronic device is configured to display a standby screen including an area displaying the face of the user avatar, and to display the first interface screen when a user command for selecting a nose portion in the face area of the standby screen is input. to,
method.
The method of claim 1,
Requesting to confirm the user's input preference value for a scent for which the verification result failed to verify the user's input preference value;
Generating the user's purchase scent history data by mapping the user's preference for scent from the user's preference scent information with the existing beauty product purchase history;
To support a purchase decision of a target beauty product having a scent, the target is based on at least a portion of the user's preference scent information including the user's preference values for a plurality of scents reflecting the verification results and the user's purchase scent history data. Calculating a matching score between the scent of a beauty product and the user; and
Further comprising: associating user-customized information with at least some of the scents among the plurality of scents,
method.
The method of claim 9, wherein the step of calculating a matching score between the scent of the target beauty product and the user comprises:
Retrieving a user's preference value for a scent related to a target beauty product from the user's preference scent information and the user's purchase scent history data; and
Comprising a step of calculating a matching score between the scent of a target beauty product and the user based on the searched user's preference value,
The matching score is characterized in that the highest ranking preference value in the preference ranking according to the user's preference value reflecting the verification result is calculated as the upper limit score,
method.
A computer-readable recording medium on which a program for performing a method for supporting user decision-making based on the scent of a beauty product in the metaverse commerce platform according to any one of claims 1 to 10 is recorded.
In a system that supports user decision-making based on the scent of beauty products on the Metaverse commerce platform,
In order to support the user's beauty product purchase decision, when receiving the user's scent information and the user's face image data, a service server is installed to verify the user's input preference value based on the user's face image data. Contains,
The user's scent information includes the user's scent preference information, including the user's scent preference information and input preference values for scents of a plurality of tasting samples,
Characterized in that the user's face image data is generated by photographing the user's face reacting to the scent while entering the user's input preference value for the scent of each tasting sample,
system.

Images