WO2023017927A1

WO2023017927A1 - Method for recommending product on basis of play type, and method for generating product recommendation model therefor

Info

Publication number: WO2023017927A1
Application number: PCT/KR2021/095078
Authority: WO
Inventors: 노희열
Original assignee: 오로라월드 주식회사
Priority date: 2021-08-11
Filing date: 2021-08-11
Publication date: 2023-02-16
Also published as: KR20230024008A

Abstract

The present disclosure provides a method for recommending a product on the basis of a play type and a method for generating a product recommendation model therefor. According to one aspect of the present disclosure, provided is a method for generating a product recommendation model by generating play type information on the basis of emotion information and/or behavior information calculated from interaction between a smart toy and a user and generating a purchaser group on the basis of product purchase information and product attribute information. According to one aspect of the present disclosure, provided is a method for recommending a product on the basis of the emotion information and/or the behavior information of a target, by means of a product recommendation model recommending a product in accordance with a play type.

Description

Product recommendation method based on play propensity and product recommendation model creation method for it

The present disclosure relates to a method for recommending a product based on play propensity and a method for generating a product recommendation model therefor. This disclosure is derived from research conducted as part of the WC300 R&D of the Ministry of Trade, Industry and Energy and the Ministry of Small and Medium Business. [Task identification number: 1425140847, task title: Interactive smart toy and expandable content service platform development through emotion recognition and interactive technology]

The contents described in this part merely provide background information on the present embodiment and do not constitute prior art.

Development of a smart toy that recognizes a user's emotion or action is increasing. Smart toys are equipped with individual sensors that detect images, voices, distances, etc. to recognize the user's emotions or behaviors calculated from interactions with the user, and provide play contents to the user based on them. The user can interact with the smart toy, obtain emotional satisfaction, and develop cognitive function and social function.

Meanwhile, in the case of toy products, since a recommended age is generally designated, a user must purchase another smart toy and build a new rapport when the smart toy he/she is using runs out of usefulness. However, in many cases, it is difficult for a user to play with a newly purchased smart toy in a conventional manner. In addition, as the user grows, the user satisfies his desire by purchasing products other than toys. However, in general, a user purchases a product that suits his or her inclination only after going through several trials and errors.

According to an aspect of the present disclosure, a product recommendation model may be generated by determining play propensity based on interaction information calculated in a process of interaction between a smart toy and a user.

According to an aspect of the present disclosure, a product may be recommended to a target based on the play propensity of the target using a pre-generated product recommendation model.

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

According to an aspect of the present disclosure, the process of collecting interaction information including user emotion information and / or user behavior information calculated in the course of interaction between the smart toy and the user; generating play propensity information based on the interaction information; Collecting product purchase information, which is product information purchased by the buyer, and product attribute information, which is attribute information about the product; generating a buyer cluster based on the product purchase information and the product attribute information; and generating a product recommendation model based on the play propensity information and the buyer cluster.

According to one aspect of the present disclosure, in the method of recommending a product to a target using a product recommendation model previously created according to the method of generating a product recommendation model according to any one of claims 1 to 7, obtaining target interaction information including emotion information and/or behavioral information of the target; and calculating a recommended product using the pre-generated product recommendation model based on the target interaction information.

According to one aspect of the present disclosure, play propensity information is generated based on emotional information and / or behavioral information calculated by interaction between a smart toy and a user, and a buyer cluster is created based on product purchase information and product attribute information to provide products By creating a recommendation model, it is possible to recommend products according to play inclinations.

According to one aspect of the present disclosure, a product may be recommended based on emotional information and/or behavioral information of a target by using a product recommendation model that recommends a product according to a play tendency.

As a result, recommended products are determined based on the target's play propensity, which has the effect of inducing the target's behavioral and emotional development in terms of play development and education.

The effects of the present disclosure 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 below.

1 is a conceptual diagram illustrating a product recommendation platform according to an embodiment of the present disclosure.

2 is a block diagram illustrating a product recommendation device according to an embodiment of the present disclosure.

3 is a flowchart illustrating a product recommendation model generation method according to an embodiment of the present disclosure.

4 is a flowchart illustrating a product recommendation method according to an embodiment of the present disclosure.

Hereinafter, some embodiments of the present disclosure will be described in detail through exemplary drawings. In adding a reference code to the components of each drawing, it should be noted that the same components have the same numbers as much as possible even if they are displayed on different drawings. In addition, in describing the present disclosure, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present disclosure, the detailed description will be omitted.

In addition, in describing the components of the present disclosure, terms such as second and first may be used. These terms are only used to distinguish the component from other components, and the nature, order, or order of the corresponding component is not limited by the term. Throughout the specification, when a part 'includes' or 'includes' a certain component, it means that it may further include other components without excluding other components unless otherwise stated. . In addition, the '... Terms such as 'unit' and 'module' refer to a unit that processes at least one function or operation, and may be implemented as hardware, software, or a combination of hardware and software.

The detailed description set forth below in conjunction with the accompanying drawings is intended to describe exemplary embodiments of the present disclosure, and is not intended to represent the only embodiments in which the present disclosure may be practiced.

A smart toy in the present disclosure refers to a toy that recognizes a user's emotion or behavior during an interaction process with the user. In the course of interaction with the user, the smart toy may acquire sensor data from at least one sensor mounted on the smart toy or mounted on a user terminal interlocked with the smart toy, and recognize the user's emotion or behavior based on the sensor data. Such sensors include, for example, a vision camera, an infrared camera, a hall sensor, a photo register, an acceleration sensor, a tilt sensor, and a magnetic sensor. It may be a magnetic switch, tact switch, touch sensor, microphone, motion sensor, gyroscope sensor, or ultrasonic sensor. , but not limited to As a method of recognizing a user's emotion or behavior based on sensor data, a means that can be easily selected by a person skilled in the art may be employed.

In the present disclosure, a toy is described as an example of a product recommended by a product recommendation platform, but the product recommendation platform of the present disclosure is not necessarily applied to be limited to these purposes or objects.

The product recommendation platform 10 according to an embodiment of the present disclosure includes all or all of a goods recommendation apparatus 100, a smart toy platform 120, and a goods sale platform 140. include some The product recommendation platform 10 shown in FIG. 1 is according to an embodiment of the present disclosure, and all components shown in FIG. 1 are not essential components, and some components may be added, changed, or deleted in other embodiments. there is.

The product recommendation device 100 collects specific data from the smart toy platform 120 and the product sales platform 140 to create a product recommendation model. The product recommendation device 100 determines a recommended product based on the user's play propensity using the generated product recommendation model. For example, the product recommendation device 100 may determine, as a recommended product, a product identical or similar to a product purchased by product buyers having the same or similar play tendency as the user, based on the user's play tendency. Such goods may be toys or smart toys, but are not limited thereto.

Specifically, the product recommendation device 100 is information calculated from the smart toy platform 120 in an interaction process between a user of a smart toy and a smart toy, and interaction information including user emotion information and/or user behavior information. can be received and collected. The product recommendation device 100 provides product information purchased by a product buyer (hereinafter, product purchase information) and/or product attribute information (eg, category, price, manufacturer, recommended age, brand, product purchase information) from the product sales platform 140. size, material, weight, shape, pattern, texture, design, etc.) may be received and collected. The product recommendation device 100 may calculate play propensity information based on the interaction information. The product recommendation device 100 may cluster buyers based on information collected from the product sales platform 140 .

The smart toy platform 120 includes all or part of a smart toy, a user interface that collects, stores, and transmits data or information detected or calculated by the smart toy, and a database. The user interface may be an app or web page mounted on the smart toy or linked with the smart toy. The user interface may work with the database to store data or information detected or calculated by the smart toy in the database or load the stored data from the database. At this time, the loaded data may be data received and stored from a plurality of user interfaces linked with the database. The smart toy platform 120 transmits interaction information between the user of the smart toy and the smart toy to the product recommendation device 100 . Interaction information includes information about smart toys (e.g., identifiers of smart toys, property information of smart toys, etc.) and information about users of smart toys (e.g., user identifiers, user personal information (gender, age, country, etc.)). More may be included.

The product selling platform 140 includes all or part of a buyer interface, a server that collects and stores purchased product information and/or product attribute information, and a database. The buyer interface may be an app or a web page for consumers to purchase products. The server may receive product purchase information, product attribute information, etc. from the buyer interface and store them in a database or load the stored data from the database. The product sales platform 140 transmits product purchase information and/or product attribute information to the product recommendation device 100 . Meanwhile, product purchase information may further include buyer information (eg, buyer identifier, buyer personal information (gender, age, country, etc.)) as well as details of the purchased product.

The product recommendation device 100 according to an embodiment of the present disclosure includes all or part of a model generator 200, an input unit 220, and a goods recommendation model 240. The product recommendation device 100 shown in FIG. 2 is according to an embodiment of the present disclosure, and all components shown in FIG. 2 are not essential components, and some components may be added, changed, or deleted in other embodiments. there is. For example, in another embodiment, the product recommendation device may further include a database (not shown) storing all or part of collected user interaction information, product purchase information, and product attribute information.

Although FIG. 2 shows the product recommendation device 100 as a device, this is for convenience of description. In another embodiment, the product recommendation device may be implemented as a software module or processor that performs the functions of each component 200 to 240. can

The model generating unit 200 collects interaction information, product purchase information, and product attribute information to create a product recommendation model. The model generator 200 according to an embodiment of the present disclosure includes a collection module 202, a play disposition analysis module 204, a clustering module 206, and a generation module ( generation module, 208), but all components shown in FIG. 2 are not essential components, and some components may be added, changed, or deleted in other embodiments.

The collection module 202 receives and collects interaction information from the smart toy platform 120 . Product attribute information, which is product purchase information and product attribute information, is received and collected from the product sales platform 140 . The collection module 202 may store the received information in a database (not shown) and load it to generate a product recommendation model. Meanwhile, information about a corresponding user and/or smart toy may be included in the interaction information.

The play propensity analysis module 204 generates play propensity information based on the interaction information. Since the interaction information includes each user's emotional and/or behavioral information generated by the interaction between the smart toy and the user, the play tendency analysis module 204 analyzes each user's play tendency based on this and calculates the play tendency information. can Play disposition is a potential characteristic that supports the cognitive, emotional, social, and personality aspects of a subject performing play, and means a motivation and disposition that enables the subject to enjoy play to the maximum.

The play propensity analysis module 204 may calculate play propensity information by calculating factors for classifying the play propensity based on the interaction information and classifying each user into a predetermined play propensity based on the values of the factors. there is. For example, the play propensity analysis module 204 determines the flexibility and agility of each user's motion based on the interaction information, the range of coordinated use between body parts, satisfaction, smile, monologue, display, mobility of the play group, curiosity about other games, Play propensity may be classified by analyzing all or part of the degree of interest, degree of contact with other users or smart toys, degree of cooperation and communication as factors. Alternatively, the play propensity analysis module 204 may calculate or classify the play propensity by combining user emotion information and user behavior information as interaction information. In this case, the play propensity analysis module 204 may use a preset play propensity table.

The clustering module 206 creates a buyer cluster based on product purchase information and product attribute information. The clustering module 206 may generate a buyer cluster by calculating a similarity between buyers based on product purchase information and product attribute information. This degree of similarity may be calculated by calculating a distance between buyers. The clustering module 206 may calculate a similarity or distance between products using each attribute included in the product attribute information as a feature, and generate a buyer cluster based on the similarity or distance between products and the product purchase information. The clustering module 206 may generate at least one buyer cluster by calculating distances between buyers for all buyers based on similarities or distances between products and classifying buyers having a short distance into one cluster. In this way, the clustering module 206 may generate a buyer cluster composed of buyers having a similar property of the purchased product and, furthermore, a similar context of the purchased product.

The generating module 208 generates the product recommendation model 240 based on the play propensity information generated by the play propensity analysis module 204 and the buyer cluster generated by the clustering module 206 . The generation module 208 may generate the product recommendation model 240 by matching the play propensity information and the buyer cluster. The generation module 208 selects a buyer cluster to which each smart toy user belongs or a buyer cluster to which buyers who have similar personal information to the corresponding user belong, and matches the play pattern information of each smart toy user. Alternatively, the generation module 208 may match the play propensity classified as play propensity information with the buyer cluster. The generating module 208 may calculate similarities between smart toy users belonging to the classified play propensity and buyers belonging to the buyer cluster, and match the classified play propensity with the buyer cluster. The generation module 208 matches the play propensity information with the buyer cluster, which is not limited to the above-described method, and means that can be selected by a person skilled in the art may be employed.

In order to recommend a product to a target using the product recommendation model 240 generated by the model generator 200, the input unit 220 receives target interaction information including emotional information and/or behavioral information of the target. Acquire The input unit 220 may receive and obtain target interaction information calculated in the course of the target interacting with the smart toy. Alternatively, the input unit 220 may obtain target interaction information by directly receiving target interaction information or by loading target interaction information pre-stored in a database (not shown).

The product recommendation model 240 calculates recommended products based on target interaction information. The product recommendation model 240 may calculate the play propensity of the target based on the target interaction information, and select a buyer cluster corresponding to the calculated play propensity. The product recommendation model 240 may calculate recommended products based on the selected buyer cluster. The product recommendation model 240 may calculate recommended products based on product attribute information corresponding to the selected buyer cluster or corresponding products.

The product recommendation device collects interaction information calculated during the interaction process between the smart toy and the user (S300). The product recommendation device may collect interaction information including emotional information and/or behavioral information of the smart toy user. The interaction information may be information calculated based on sensor data sensed by the smart toy.

The product recommendation device calculates play propensity based on the interaction information and generates play propensity information (S302).

The product recommendation device collects product purchase information, which is product information purchased by the buyer, and product attribute information, which is attribute information about the product (S304).

The product recommendation device calculates the similarity between products based on the product attribute information and creates a buyer cluster based on the similarity between products and the product purchase information (S306). The product recommendation device may calculate a similarity between products by calculating a distance between products based on a difference between product attribute information of each product.

The product recommendation device generates a product recommendation model by matching the play propensity information with the buyer cluster (S308).

The product recommendation device acquires target interaction information including target emotion information and/or target behavior information (S400).

The product recommendation device calculates the play propensity of the target based on the target interaction information (S402).

The product recommendation device selects and obtains a buyer cluster corresponding to the calculated play propensity (S404).

The product recommendation device determines a recommended product based on the acquired product attribute information and/or product corresponding to the buyer cluster (S406).

In FIGS. 3 and 4, it is described that each process is sequentially executed, but this is merely an example of the technical idea of an embodiment of the present disclosure. In other words, those skilled in the art to which an embodiment of the present disclosure belongs may change and execute the order described in FIGS. 3 and 4 without departing from the essential characteristics of an embodiment of the present disclosure, or one of each process Since it will be possible to apply various modifications and variations by executing the above process in parallel, it is not limited to the time-sequential order of FIGS. 3 and 4.

Various implementations of devices, units, processes, steps, etc. described herein may include digital electronic circuits, integrated circuits, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), computer hardware, firmware, software, and/or combinations thereof. These various implementations may include being implemented as one or more computer programs executable on a programmable system. A programmable system includes at least one programmable processor (which may be a special purpose processor) coupled to receive data and instructions from and transmit data and instructions to a storage system, at least one input device, and at least one output device. or may be a general-purpose processor). Computer programs (also known as programs, software, software applications or code) contain instructions for a programmable processor and are stored on a “computer readable medium”.

A computer-readable recording medium includes all kinds of recording devices that store data that can be read by a computer system. These computer-readable 　recording media include non-volatile or non-transitory storage devices such as ROM, CD-ROM, magnetic tape, floppy disk, memory card, hard disk, magneto-optical disk, and storage device. It may further include a medium. In addition, the computer-readable recording medium may be distributed in computer systems connected through a network, and computer-readable codes may be stored and executed in a distributed manner.

Various implementations of the systems and techniques described herein may be implemented by a programmable computer. Here, the computer includes a programmable processor, a data storage system (including volatile memory, non-volatile memory, or other types of storage systems, or combinations thereof) and at least one communication interface. For example, a programmable computer may be one of a server, network device, set top box, embedded device, computer expansion module, personal computer, laptop, personal data assistant (PDA), cloud computing system, or mobile device.

The above description is merely an example of the technical idea of the present embodiment, and various modifications and variations can be made to those skilled in the art without departing from the essential characteristics of the present embodiment. Therefore, the present embodiments are not intended to limit the technical idea of the present embodiment, but to explain, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of this embodiment should be construed according to the claims below, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of rights of this embodiment.

10: Product recommendation platform

<CROSS-REFERENCE TO RELATED APPLICATION>

This patent application claims priority to Patent Application No. 10-2021-0105998 filed in Korea on August 11, 2021, all of which are incorporated into this patent application by reference.

Claims

Collecting interaction information including user emotion information and/or user behavior information calculated in the course of interaction between the smart toy and the user;

generating play propensity information based on the interaction information;

Collecting product purchase information, which is product information purchased by the buyer, and product attribute information, which is attribute information about the product;

generating a buyer cluster based on the product purchase information and the product attribute information; and

Process of generating a product recommendation model based on the play propensity information and the buyer cluster

Characterized in that it comprises a, product recommendation model generation method.
According to claim 1,

The interaction information is calculated based on sensor data detected from all or part of a camera, a thermal imaging camera, and a gyroscope sensor mounted in the smart toy.
According to claim 1,

The process of generating the play propensity information,

The method of generating a product recommendation model, characterized in that generating the play propensity information by calculating the play propensity by combining the user emotion information and the user behavior information.
According to claim 1,

The process of creating the buyer cluster,

The method of generating the product recommendation model, characterized in that the buyer cluster is generated by calculating a similarity between buyers based on the product purchase information and the product attribute information.
According to claim 4,

The process of creating the buyer cluster,

The method of generating a product recommendation model, characterized in that calculating a similarity between products based on the product attribute information, and generating the buyer cluster based on the similarity between products and the product purchase information.
According to claim 1,

The process of generating the product recommendation model,

The method of generating the product recommendation model, characterized in that the product recommendation model is generated by matching the play propensity information with the buyer cluster.
According to claim 6,

The process of generating the product recommendation model,

The method of generating a product recommendation model, characterized in that selecting a buyer cluster corresponding to the user from among the buyer clusters, and matching the selected buyer cluster with the play pattern information.
A computer program stored in a computer-readable recording medium to execute each process included in the product recommendation model generation method according to any one of claims 1 to 7.
In the method of recommending a product to a target using a product recommendation model previously generated according to the method of generating a product recommendation model according to any one of claims 1 to 7,

obtaining target interaction information including emotion information and/or behavioral information of the target; and

Calculating a recommended product using the pre-generated product recommendation model based on the target interaction information

Characterized in that, a product recommendation method comprising a.
According to claim 9,

The process of calculating the recommended product,

The method of recommending a product, characterized in that using the product recommendation model, calculating the play propensity of the target based on the target interaction information, and selecting a buyer cluster corresponding to the calculated play propensity.
According to claim 10,

The process of calculating the recommended product,

The product recommendation method characterized in that the product recommendation method is determined and calculated based on product attribute information corresponding to the selected buyer cluster using the product recommendation model.
According to claim 10,

The process of calculating the recommended product,

The method of recommending a product, characterized in that determining and calculating the recommended product based on a product corresponding to the selected buyer cluster using the product recommendation model.