KR101846422B1 - Clothing recommendation system considering individual characteristics and preferences - Google Patents

Abstract

A customized online and offline linked clothing recommendation system in consideration of individual characteristics and preferences using big data, includes: a database for storing user information and clothing information for a user associated with a user terminal; a collection unit for collecting the clothing information from the database; a preference calculating unit for calculating the preference of the user based on preference survey result information and purchase history information included in the user information; an activity index calculating unit for calculating a clothing activity index based on clothing dimension information and clothing material information included in the clothing information; and a recommendation unit for determining recommendation clothing based on the preference of the user, the user activity index included in the user information, and the clothing activity index. Accordingly, the present invention can recommend customized clothing by considering the individual characteristics and preferences.

Description

[0001] CLOTHING RECOMMENDATION SYSTEM CONSIDERING INDIVIDUAL CHARACTERISTICS AND PREFERENCES [0002] BACKGROUND OF THE INVENTION [0003]

The present invention relates to a garment recommending system, and more particularly, to a garment recommending system that recommends a customized garment in consideration of personal characteristics and preferences by utilizing big data, and links online and offline.

The present invention relates to a clothing recommendation system and a recommendation method for recommending information on an actual apparel product corresponding to each user's taste to the user by analyzing the apparel purchase taste of each user using the avatar apparel purchase information of users.

Recently, the number of Internet users who purchase goods using e-commerce has been increasing due to the development of the Internet. As the Internet shopping mall has developed as an important means of commerce, the number of Internet shopping malls handling clothes is also increasing and the number of products handled is increasing.

Currently, clothing is displayed and sold in the Internet shopping mall by classifying the clothes according to the brand or the type, and the image information and the detailed description (product code, brand name, color, material, manufacturer, ) Are exposed on an Internet site such as a web site, and uniform and passive methods are awaiting the selection of the Internet user. Also, a method of exposing clothing information based on clothing items purchased by many people is also used, but this refers to collective purchasing records and does not reflect the individualized taste of each individual.

Korean Laid-Open Patent No. 2017-0021454 (Published Feb. 28, 2017)

An object of the present invention is to provide a clothing recommendation system which can recommend a customized clothing considering individual characteristics and preferences.

It is an object of the present invention to provide a clothing recommendation system capable of linking an online store and an offline store.

In order to accomplish one object of the present invention, a customer-oriented on-off-line linked garment recommendation system that takes individual characteristics and preferences into consideration by utilizing big data according to embodiments of the present invention, A database for storing clothing information; A collection unit for collecting clothing information from the database; A preference calculating unit for calculating user preference based on the preference questionnaire result information and purchase history information included in the user information; An activity index calculating unit for calculating an apparel activity index based on the apparel dimension information and the apparel material information included in the apparel information; And a recommendation unit for determining a recommendation garment based on the user preference, the user activity index included in the user information, and the clothing activity index.

According to one embodiment, the activity index calculating unit may calculate the apparel activity index based on the following equation (1).

[Equation 1]

AI (activity index) = S_k / SR_k * W1_k + W2_k *? Mi * Ri

Here, AI is a clothing activity index, k is a clothing identification number, S_k is a circumference length of the garment, SR_k is a reference circumference length of a clothing group (for example, a small classification CLASS3) W1_k is the first weight of the garment, W2_k is the second weight of the garment, Mi is the activity index of the ith material of the garment, and Ri is the ratio of the ith material of the garment.

According to one embodiment, the collecting unit collects the activity information of the user from the user terminal, the activity information includes the user position information of the user by time, and the activity index calculating unit calculates the activity index, Can be calculated.

According to an embodiment, the activity index calculating unit may include a speed calculating unit that calculates a moving speed with time based on the user position information, and generates a first moving speed graph indicating a change in the moving speed with time, ; A speed correcting unit for generating a second moving speed graph by removing a section of the first moving speed graph exceeding a first reference speed; And a calculation unit that calculates the user activity index by integrating the second moving speed graph based on the area weight set for each real area to which the position of the user belongs.

According to an embodiment of the present invention, the recommendation unit searches the recommended offline store for selling the recommendation clothing based on the offline store information included in the database, and searches the recommended offline store based on the current location information of the user terminal and the address information of the user It is possible to determine the recommended offline store.

According to an embodiment of the present invention, the garment recommendation system generates price change information according to time based on price information of the recommendation garment included in the database, And a price analyzing unit for correlating the event information and generating the price change information.

According to an embodiment, the price analyzing unit may classify the price information of the recommendation garment. In this case, the recommendation unit may display the recommended offline store together with the graded price information on the map.

The on-line and off-line linked clothing recommendation system, which takes advantage of the big data according to the embodiments of the present invention and considers individual characteristics and preferences, determines the preference and the preference goods of the user using the survey information, The apparel activity index, and the user activity index are calculated based on the user information (the user's position information, the user's position information, and the user activity information) It is possible to more objectively determine the user's preference and the user's characteristic by determining the recommendation garment for the user based on the recommendation of the user, and to recommend the garment suitable for the user, that is, the user-customized garment to the user.

In addition, the customized on / off-line linked clothing recommendation system that utilizes the big data to take individual characteristics and preferences into account, determines a replacement product and / or supplementary product based on the user and / or product, By displaying price rating information and displaying the store information on a map, the user can select a variety of user-customized products, intuitively determine the purchase timing of the products, the offline purchase location of the products, Activation of offline stores may be possible.

However, the effects of the present invention are not limited to the above effects, and may be variously extended without departing from the spirit and scope of the present invention.

1 is a block diagram illustrating a garment recommendation system in accordance with embodiments of the present invention.
FIG. 2 is a block diagram showing an example of a clothing recommendation server included in the clothing recommendation system of FIG. 1. FIG.
FIG. 3A is a block diagram illustrating an example of a preference analyzer included in the recommendation server of FIG. 2; FIG.
FIG. 3B is a diagram illustrating an example of a preference questionnaire provided by the preference analyzer of FIG. 3A.
FIG. 3C is a diagram showing an example of the preferred clothing list determined by the preference analyzing unit of FIG. 3A. FIG.
FIG. 4A is a diagram showing an example of clothing information used in the clothing recommendation server of FIG. 2. FIG.
FIG. 4B is a diagram showing an example of user information used in the recommendation server of FIG. 2; FIG.
5 is a block diagram showing an example of an activity index calculating unit included in the garment recommendation server of FIG.
6A to 6C are diagrams illustrating a process of calculating a user activity index in the activity index calculating unit of FIG.
FIG. 7 is a block diagram showing an example of a recommendation unit included in the recommendation server of FIG. 2; FIG.
8A to 8D are views showing an example of recommended clothing information generated by the recommendation unit of FIG.
9 is a flowchart showing a recommended garment providing method performed in the recommended garment system of Fig.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof. Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a garment recommendation system in accordance with embodiments of the present invention.

Referring to FIG. 1, a customized on / off line linked clothing recommendation system 100 (hereinafter, referred to as a "clothing recommendation system") that takes individual characteristics and preferences into consideration by utilizing big data includes a user terminal 110, (120). Here, the user terminal 110 and the clothes recommendation server 120 may be interconnected via a wired / wireless network.

The user terminal 110 receives a product search request from a user (or a customer) possessing or using the user terminal 110 through an input module (for example, a touch screen, a microphone, etc.) Transmits the goods search request signal to the clothing recommendation server 120 and provides the recommendation clothing information (or the recommended clothing list) provided by the clothing recommendation server 120 to the user through an output module (for example, a display) . For example, the user terminal 110 may be a desktop, laptop, tablet, smart phone, or the like. When the user terminal 110 is implemented as a smartphone, the location information of the user is acquired through a GPS module embedded in the smart phone, and the user's activity information (eg, For example, the number of steps). The acquired location information and / or activity information of the user can be used to calculate the user activity index in the recommendation server 120. [ Alternatively, when the user terminal 110 is implemented as a desktop, the user terminal 110 may be associated with a device (e.g., smart phone, pedometer, etc.) worn by the user, Can be obtained.

The garment recommending server 120 may analyze the big data associated with the user and provide clothing information (or a list of recommended garments) suitable for the user to the user terminal 110. [ For example, the apparel recommendation server 120 may store a user's characteristics (e.g., a user's activity) and preferences (e.g., user preferences) based on user information (e.g., location information, And the clothes matching the user's characteristics and preferences can be determined as the recommended clothes. In addition, the clothing recommendation server 120 may provide relevant information (e.g., inventory, location information, event information, etc.) of the offline store for selling the recommended clothing together with the recommended clothing information. That is, the clothing recommendation server 120 can provide information suitable for various users such as a cross shopper (for example, omni-shopper) by linking an online store and an offline store.

The specific configuration of the clothes recommendation server 120 will be described later with reference to Fig.

Meanwhile, the garment recommendation system 100 may include a seller server 130, an activity amount analysis server (not shown), or the like, or may be connected to them through a network.

The seller server 130 may include information on the seller of the seller who sells the clothes, information on the clothes, inventory, etc. sold by the seller. For example, the merchant server 130 may be an online store.

The activity analysis server analyzes the activity amount of the user based on the user information (for example, the user's position information, the number of steps of the user, the movement speed of the user, etc.). For example, Or may be implemented as a server or program.

As described above, the garment recommending system 100 can provide / recommend a user-customized garment (or a custom-made garment) by calculating / determining the characteristics and preferences of the user and selecting / determining the recommended garment. In addition, the apparel recommendation system 100 can provide the store information of the offline store that sells the recommended apparel along with the recommended apparel information, thereby linking the online store with the offline store and providing customized information to various types of users .

Meanwhile, the garment recommending system 100 may be implemented as a recording medium storing a program or a program including the functions described above, and installed / operated in the user terminal 110, the clothing recommendation server 120, and the like.

FIG. 2 is a block diagram showing an example of a clothing recommendation server included in the clothing recommendation system of FIG. 1. FIG.

2, the recommended clothing server 120 includes a communication unit 210, a collection unit 220, a preference degree analysis unit 230, an activity index calculation unit 240, an input / output unit 250, a recommendation unit 260, A database 270, a price analyzing unit 280, and a control unit 290.

The communication unit 210 receives the user information from the user terminal 110, transmits the recommended clothing information to the user terminal 120, transmits and receives the clothes information to and from the seller server 130, And can receive user activity information. For example, the communication unit 210 may be implemented as a communication module using a wire / wireless communication method.

The collecting unit 220 may collect user information and apparel information. Herein, the user information includes user basic information such as the user's ID, name, sex, age, address, etc., user purchasing history (for example, history of purchase of clothing by the user during a specific period of time), user preference (E.g., preference item information, preference questionnaire information, etc.), location information of the user (e.g., location information of the user's hourly time), user's activity information have. Also, the user information may include user group information (for example, a user group to which the user belongs among user groups classified based on the preference survey result) generated by the preference analyzing unit 210, activity index calculating unit 240 And the user activity index generated by the user. Similarly, the clothing information includes clothing basic information such as clothing code, clothing classification, dimensional information, and the like, clothing activity index calculated by the activity index calculating unit 240, in-store information for selling / holding the clothing, , Store event information, and the like.

For example, the collecting unit 220 requests user information, clothing information, and the like to the user terminal 110, the seller server 130, and the database 270, and transmits the received information to the recommendation server 120 (Or the garment recommending system 100). The collected user information and clothing information may be stored / updated in the database 270.

The preference analyzing unit 230 can calculate the user's preference based on the user information.

In one embodiment, the preference analyzer 230 may calculate the user's preference based on the preference survey result (or survey result information) and purchase history information.

Hereinafter, the preference analyzing unit 230 will be described in detail with reference to FIGS. 3A to 3C, and the activity index calculating unit 240 and the like will be described.

FIG. 3A is a block diagram illustrating an example of a preference analyzer included in the recommendation server of FIG. 2, FIG. 3B is a diagram illustrating an example of a preference questionnaire provided by the preference analyzer of FIG. 3A is a diagram showing an example of the preferred clothing list determined by the preference analyzing unit of FIG.

3A, the preference analyzing unit 230 may include a surveying unit 310, a purchase history analyzing unit 320, and a preferred goods determining unit 330.

The surveying unit 310 generates surveying information based on the clothing classification included in the clothing information, provides the surveying information to the user terminal 110 through the communication unit 210, and based on the survey result, (I.e., the user who owns / uses the user terminal 110 as the user associated with the user terminal 110).

For example, the survey unit 310 belongs to the same category (e.g., the middle class CLASS2 shown in FIG. 4A) as shown in FIG. 3B, and other attributes (for example, (CLASS3), or Clothes Activity Index (AI)) to generate survey information. For example, the surveying unit 310 may include a first sub-questionnaire for requesting selection of one of four products that correspond to " pants " of the middle class " CLASS2 " Information can be generated. Similarly, the survey unit 310 may generate the second sub-survey information by arbitrarily selecting three products having the activity index (AI) corresponding to the " T-shirt " in the middle class CLASS2. The surveying unit 310 may repeat this procedure to generate N sub-survey inquiry information (where N is an integer of 2 or more). N sub-survey information may be included in the survey information.

In this case, the survey information may be provided to the user terminal 110, and the survey result information may be provided through the user terminal 110. For example, the apparel recommendation server 120 may provide survey information at the initial access / connection of the user to the apparel recommendation server 120, or may be provided to the user terminal 110 at the time of generating the survey information Survey information can be provided. The N sub-survey inquiry information may be sequentially provided to the user through the user terminal 110. In response to this, N sub-survey inquiry information (for example, product information selected corresponding to the sub- ) Can be sequentially acquired / generated. N sub-survey result information can be included in the survey result information.

The surveying unit 310 can determine the user group to which the user belongs based on the survey result information. For example, the survey unit 310 may set a plurality of user groups based on the type (or type) of the expected survey result information, and determine a group of users matching the survey result information of the user , A group of users having the same or similar survey result information).

Meanwhile, the survey unit 310 generates a plurality of survey information based on the user's sex, age, and the like, selects survey information corresponding to the user's sex, age, )).

The purchasing history analyzing unit 320 may analyze the purchasing history of each user to extract preferred clothes from all the clothes. For example, the purchasing history analyzing unit 320 may search the clothes having history that the users purchase based on the purchase history included in the user information of the users belonging to the specific user group, have.

The preferred product determination unit 330 can determine the preferred goods (or preferred clothing) for each user group based on the preferred clothing (or the preferred clothing list). For example, the preferred merchandise determination unit 330 rearranges the preferred merchandise based on the sales quantity and the evaluation score of the preferred clothing, and determines M (M is a positive integer) having the highest ranking as the preferred merchandise . The preferred product deciding section 330 can determine a preferred commodity for each of the major classification CLASS1, the middle classification CLASS2 and the small classification CLASS3 shown in FIG. 3C.

As described with reference to FIGS. 3A to 3C, the preference analyzing unit 230 generates survey information based on basic information (for example, sex and age) of the user and the type of clothing, The user group of the user is determined based on the survey result corresponding to the survey information, and the preference item for the user (and the user group) is determined through the purchase history analysis of the user group have.

Referring again to Fig. 2, the activity index calculating unit 240 may calculate the apparel activity index based on the apparel information, and calculate the user activity index based on the user information. Here, the clothing activity index indicates the degree of suitability of the user wearing the garment, and can be quantitatively expressed by integrating the space / width of the garment, the permeability of the material of the garment, and the weight. Similarly, the user activity index indicates the amount of physical activity of the user, and can be expressed in a form of a sum of the user's activity information such as the moving distance, the moving speed, the number of steps, etc. of the user.

Hereinafter, the activity index calculating unit 240 will be described in detail with reference to FIGS. 4A to 6C, and the input output 250 and the like will be described.

FIG. 4A is a diagram showing an example of clothing information used in the recommendation server of FIG. 2, FIG. 4B is a diagram showing an example of user information used in the recommendation server of FIG. 2, 6A to 6C are diagrams illustrating a process of calculating a user activity index in the activity index calculating unit of FIG. 5; FIG.

The activity index calculating unit 240 may calculate the garment activity index based on the garment dimension information and the garment material information included in the garment information.

Referring to FIG. 4A, the clothing information includes clothing code (or clothing identification number), clothing classification information CLASS1, CLASS2, CLASS3, clothing size information SIZE_C, SIZE_LENGTH, SIZE_WAIST, SIZE_HIP, SIZE_HEM, Weight information (WEIGHT), clothing material information (FABRIC1, FABRIC2, FABRIC3), clothing activity index (AI), and the like. The garment classification information CLASS1, CLASS2 and CLASS3, that is, the major classification CLASS1, the middle classification CLASS2 and the minor classification CLASS3 are classified into three types of garment types or types, , This is an example, and the clothing classification information is not limited thereto. For example, the garment classification information may include sub-categories of one, two, or four or more levels. In addition, the clothing information may include product information of a complementary product (e.g., a product such as shoes) worn by the user along with the clothing in addition to the clothing.

The garment dimension information SIZE_C, SIZE_LENGTH, SIZE_WAIST, SIZE_HIP, and SIZE_HEM includes the size or size of the garment. For example, the first garment dimension information SIZE_C represents the total size of the garment, The information SIZE_LENGTH indicates the length of the garment, the third garment dimension information SIZE_WAIST or SIZE_CHEST indicates the circumference length of the garment, and the fourth garment dimension information SIZE_HIP or SLEEVE LENGTH indicates the size (For example, the size of the bottom portion in case of the bottom and the length of the sleeve in the case of the top), and the fifth garment dimension information SIZE_HEM may indicate the length of the edge of the garment (e.g., the bottom). This is an example, and the garment dimension information is not limited thereto.

For example, the first garment material information FABRIC1 indicates the ratio of the first garment material information FABRIC1 to the first garment material information FABRIC2, The second garment material information FABRIC2 may represent the ratio of the second material (e.g., NYLON), and the third garment material information FABRIC3 may represent the ratio of the third material (e.g., POLY). The clothing material information FABRIC1, FABRIC2, and FABRIC3 includes three pieces of clothing material information, but this is exemplary. The clothing material information FABRIC1, FABRIC2, and FABRIC3 includes information on four or more materials can do.

In one embodiment, the activity index calculating section 240 may calculate the apparel activity index based on the following equation (1).

[Equation 1]

AI (activity index) = S_k / SR_k * W1_k + W2_k *? Mi * Ri

Here, AI is a clothing activity index, k is a clothing identification number, S_k is a circumference length of the garment (for example, third garment dimension information SIZE_WAIST or SIZE_CHEST), SR_k is a clothing group W1_k is the first weight of the garment, W2_k is the second weight of the garment, Mi is the activity index of the i-th material of the garment, Ri is the weight of the first garment, I < / RTI > material of the garment. The first weight W1_k and the second weight W2_k may be predefined for each of the clothing classification information CLASS1, CLASS2 and CLASS3.

On the other hand, the activity index calculating unit 240 may calculate the activity index based on the weight information WEIGHT, the fabric thickness information (not shown) (for example, fabric thickness such as 10, 40, and the like) The apparel activity index (AI) can be calculated.

Meanwhile, the activity index calculating unit 240 may calculate the user activity index based on the user activity information. Here, the user activity information includes user position information, user activity amount (e.g., number of steps), generated in the GPS module / gyro module built in the user terminal 110, (Not shown). Meanwhile, the user activity index may be included in the user information as shown in FIG. 4B.

5, the activity index calculation unit 240 may include a speed calculation unit 510, a speed correction unit 520, and a calculation unit 530. [

The speed calculating unit 510 may calculate the moving speed according to the time (or the time variation) based on the user position information, and may generate the first moving speed graph showing the variation of the moving speed with the time variation.

As shown in FIG. 6A, the user location information may be displayed on the map MAP1 as the user movement route ROUTE1 according to the change of time. Here, the map MAP1 can be obtained from a map providing server (not shown). For example, the first area AREA1 represents a residential area, the second area AREA2 represents a commercial area, and each area weight (for example, a user A weight used to calculate the activity index). The first and second areas AREA2 are illustrative and the regional usage information can be further subdivided, such as a central commercial area, a general commercial area, a neighborhood commercial area, a commercial commercial area, and the like.

Referring to FIG. 6B, there is shown a first moving speed graph G1 showing a change in the moving speed of the user with time variation. The speed calculating unit 510 calculates the moving speed based on the change of the user position information per unit time, and can display the change of the moving speed in a specific time (for example, one day).

Meanwhile, the speed corrector 520 may generate a second moving speed graph by removing a section of the first moving speed graph that exceeds the first reference speed. Here, the first reference speed may correspond to the maximum speed of the user (or person). That is, the speed corrector 520 may determine that a section exceeding the first reference speed is generated through a moving means such as an automobile, not a user activity (for example, a walk), and may delete the section. Through this, the speed corrector 520 can generate the second traveling speed graph G2 as shown in FIG. 6B.

Meanwhile, although the first reference speed is described as being the maximum speed of the user, this is merely exemplary and the first reference speed is not limited thereto. For example, the first reference speed may be the maximum speed of the non-powered moving means, such as a bicycle.

The calculating unit 530 may calculate the user activity index based on the weight and the second moving velocity graph G2 set for the actual region to which the user's position belongs.

For example, the calculating unit 530 calculates the second moving speed graph G2 (AREA2) based on the area weights predefined for each area (for example, the first area AREA1 and the second area AREA2 shown in FIG. 6A) And the user activity index can be calculated by normalizing the integrated second movement speed graph G2 (for example, generalizing using the normal distribution function).

For example, the calculating unit 530 may divide the second moving velocity graph G2 into a plurality of intervals P1, P2, and P3 based on the user position information, integrate the moving velocity on each of the intervals, The user activity index can be calculated through a weight calculation that multiplies the moving speed (or re-calculated distance) by the area weight.

In one embodiment, the calculating unit 530 may calculate the user activity index based on the user activity information.

Referring to FIG. 6C, a third moving speed graph G3 showing the number of steps of the user in accordance with a time change is shown. Here, the number of steps may be acquired / generated from the user terminal 110 (or a pedometer linked to the user terminal 110) described above with reference to FIG. 1 and provided to the recommendation server 120.

For example, the calculator 510 calculates / determines the user's stride on the basis of the change of the user's position information and the number of steps of the user in a specific section, multiplies the stride of the user by the total number of steps of the user The activity index of the user can be calculated. As described above with reference to FIG. 6B, the calculating unit 530 may calculate the user activity index by using the area weight or by generalizing using a normal distribution function or the like.

On the other hand, the calculating unit 510 can calculate the activity index for each of a plurality of periods. 4B, the calculating unit 510 calculates the first activity index (or activity index 1) in units of one day, and calculates the first activity index during the first specific period unit such as weekday or working period The second activity index (or activity index 2) can be calculated on the average. Similarly, the calculating unit 510 calculates a third activity index (or an activity index 3) by averaging the first activity index during a second specific time period such as a weekend or a holiday, 4 Activity index (or activity index 4) can be calculated. The calculated plurality of user activity indexes (i.e., the first to fourth activity indexes) may be used to determine a specific type of recommended garment. For example, since it is common for a shirt and pants of a male in his thirties to be worn by a user during a week (or a working time), the garment recommendation server 120 may search the corresponding user For men in their 30s, you can decide on a recommended garment for.

As described with reference to FIGS. 5 to 6C, the activity index calculating unit 240 may calculate the user activity index using user location information, user activity information, and the like.

2, the input / output unit 250 receives setting information such as the first and second weights W1 and W2, the area weights, and the like for each garment from the manager, and outputs the calculated clothing activity index, user activity index, Can be provided to the user. The input / output unit 250 may be implemented as a general input / output module such as a keyboard, a mouse, and a display device.

The database 270 may store the preferred clothing information described with reference to FIG. 3C, the clothing information described with reference to FIG. 4A, and the user information described with reference to FIG. 4B. The database 270 may be implemented as a storage device such as a memory device, or may be implemented as a separate server.

The recommendation unit 260 can determine the recommended clothing based on the user preference, the user activity index, and the clothing activity index.

In one embodiment, the recommendation unit 260 can determine the recommendation garment based on the user preference (i.e., the preference item determined by the preference analysis unit 230). For example, when the user searches for " pants ", one of the preferred clothes determined based on the user group to which the user belongs can be determined as the recommended clothes corresponding to " prevention "

For reference, it may be that the garments purchased by a user group having a similar age of a user or purchased by other users who bought the same clothing in the past may be considered only the general characteristics of the user. Alternatively, the garment recommending system 100 (or the garment recommending server 120) may increase the objectivity of determining the user's preference (or setting of the user group) by using the survey information, It is possible to provide a recommendation product suitable for the user.

In one embodiment, the recommendation unit 260 may determine garments having a clothing activity index matching the user activity index as the recommended garment.

For example, the garment activity index described with reference to FIG. 4A has a value of a specific interval (for example, 0 to 100) through a generalization process, and similarly, the user activity index described with reference to FIG. (E. G., 0 to 100). ≪ / RTI > In this case, the recommendation unit 260 may search for garments having the same or similar clothing activity index (for example, 70 or 65 to 75) as the user activity index of the user (for example, as the third activity index, It can be decided by recommended clothing. In addition, the recommendation unit 260 rearranges the recommended garments based on the sales volume and the evaluation score for each recommended garment, generates a recommended garment list as shown in FIG. 8A, and provides it to the user through the user terminal 110 .

Meanwhile, the price analyzing unit 280 generates price change information according to time based on the price information of the recommended garment included in the database 270, associates the event information of the recommended in-store corresponding to the price change point, Change information can be generated. The specific structure of the price analysis unit 280 will be described later with reference to Figs. 8C and 8D.

The control unit 290 includes a communication unit 210, a collecting unit 220, a preference analyzing unit 230, an activity index calculating unit 240, an inputting and outputting unit 250, a recommending unit 260, a database 270, The operation of each of the units 280 and the data (or the flow of data) between them can be controlled.

As described with reference to FIGS. 2 to 6C, the clothing recommendation server 120 (or the clothing recommendation system 100) determines the user's preference and preferred goods using the survey information, Based on the user's preference, the clothing activity index, and the user activity index, based on the user's information (e.g., the size information of the clothes, the size information of the clothes, and the clothes material information) To determine a recommended garment for the user. Accordingly, the apparel recommendation server 120 (or the apparel recommendation system 100) can more objectively determine the user's preference and the user's characteristics, and recommend the apparel suitable for the user, that is, the user-customized apparel to the user .

FIG. 7 is a block diagram showing an example of a recommendation unit included in the recommendation server of FIG. 2, and FIGS. 8A to 8D are diagrams showing an example of recommended recommendation information generated by the recommendation unit of FIG.

Referring to FIGS. 2 and 7, the recommendation unit 260 may include an alternative product recommendation unit 710 and a supplementary product recommendation unit 720.

The substitute product recommendation unit 710 can determine the recommended apparel belonging to the same category as the apparel searched by the user. For example, when the user searches for " pants "in the middle class (CLASS2) shown in FIG. 4A, it is possible to determine the recommended clothes belonging to " pants "

In an embodiment, the substitute product recommendation unit 710 may include a first recommendation unit 711 and a second recommendation unit 712.

The first recommendation unit 711 can determine the first recommended product (or the first recommended product) based on the user information. For example, the first recommendation unit 711 may determine the preferred product determined based on the user group to which the user belongs as the first recommended product. For example, the first recommendation unit 711 may determine that the user activity index of the user is equal to or similar to other users (for example, if the user activity index of the user is 70, the user activity index within the range of 65 to 75 The first recommendation commodity can be determined based on the purchasing history of other users. The configuration for determining the first recommended product based on the purchasing history of the other users is substantially the same as the configuration for determining the preferred product described with reference to FIG. 3A, so that redundant description will not be repeated.

That is, the first recommendation unit 711 can determine the first recommendation commodity based on the user's preference, the same / similar affinity with the user activity index, and another user having the user activity index (i.e., similar user).

The second recommendation unit 712 can determine the second recommended product (or the second recommended product) based on the clothing information. For example, the second recommendation unit 712 may search for a product having an apparel activity index that is the same as or similar to the user activity index of the user, and may determine the retrieved product as the second recommendation product.

That is, the second recommendation unit 712 can determine the second recommendation commodity based on the commodity having the apparel activity index similar to the user activity index.

On the other hand, the supplementary product recommendation unit 720 can determine a recommended product belonging to a category different from the clothing searched by the user. For example, when the user searches for " pants "in the middle class (CLASS2) shown in FIG. 4A, a recommendation garment belonging to " (For example, shoe) belonging to the category can be determined as a recommended product. Here, the type of the supplementary product is predetermined, and for example, a plurality of reference users (for example, When receiving a search request for a specific product (for example, pants) based on the products (for example, three products such as cotton, pants, shoes, etc.) May be determined as " cottony " and " shoes ".

In one embodiment, the complementary product recommendation unit 720 may include a third recommender unit 721 and a fourth recommender unit 722.

Similar to the first recommendation unit 711, the third recommendation unit 721 can determine the third recommendation commodity based on the user information. For example, the third recommendation unit 721 may determine the preferred product determined based on the user group to which the user belongs as the third recommended product. For example, the third recommendation unit 721 may determine that the user activity index of the user is equal to or similar to other users (for example, if the user activity index of the user is 70, the user activity index within the range of 65 to 75 Based on the purchasing history of the first recommendation commodity.

That is, the third recommendation unit 721 can determine the third recommendation commodity based on the user's preference, the same / similar affinity with the user activity index, and another user having the user activity index (i.e., similar user).

Similar to the second recommendation unit 712, the fourth recommendation unit 722 can determine the fourth recommendation commodity based on the apparel information. For example, the fourth recommendation unit 722 may search for a product having a clothing activity index that is the same as or similar to the user activity index of the user, and may determine the retrieved product as the fourth recommended product.

That is, the fourth recommendation unit 722 can determine the second recommendation commodity based on the commodity having the clothing activity index similar to the user activity index.

The recommendation unit 250 may combine the first to fourth recommended products and provide the recommended product list LIST1 shown in FIG. 8A to the user through the user terminal 110. [

In embodiments, the recommendation unit 250 may generate seller information (or offline store information) associated with the referral product.

In one embodiment, the recommendation unit 250 searches the offline store for holding / selling the recommended clothing based on the offline store information included in the database 270, and stores the current location information of the user terminal 110 (E.g., current location information of the user) and user address information (e.g., home address, work address). Here, the offline store information may include offline store information such as a store address, a sale item, a sale item inventory, a price, and the like for each store as in the second offline store list LIST2 shown in FIG. 8B.

For example, the recommendation unit 250 may include a first recommended offline store (for example, the first store SHOP1 shown in FIG. 8D) located within a first radius (for example, 1 Km) ) Can be searched. For example, the recommendation unit 250 may include a second recommended offline store (e.g., a second store (e.g., a second store) located within a second radius (e.g., 1 km) based on the user's home address and / SHOP2)). In another example, the recommendation unit 250 may include a third recommended offline store (e.g., a third recommended offline store) located within a third radius (e.g., 500 m) based on a user's user movement path ROUTE1 (see FIG. 6A) , The third store (SHOP3)).

On the other hand, the recommendation unit 250 displays the position information of the recommended offline store on the map MAP1 shown in Fig. 8B, and displays a button " map view " included in the recommended goods list LIST1 shown in Fig. Corresponding to the input to the user, the recommended offline store map information shown in Fig. 8D can be provided to the user.

2 and 8C, the price analyzing unit 280 generates price change information according to time based on the price information of the recommended garment included in the database 270, and stores price information (or price) The price change information can be generated by associating the event information of the recommended offline store with the changing price change point (or section).

Referring to FIG. 8C, the price of the recommended garment (or recommended product) can be shown according to the change of time. The first pricing line PG1 represents the change in the maximum price of the recommended garment and the second price line PG2 represents the change in the minimum price of the recommended garment.

When the price change information includes a period in which the price of the recommended garment is changed, such as the second price line PG2, the price analyzing unit 280 stores the event information corresponding to the event corresponding section in the offline store information (see FIG. 8B) And associate the retrieved event information with the corresponding section. For example, event information (or discount information) of the recommended offline store (for example, the first offline store SHOP1) can be displayed corresponding to the section.

In one embodiment, the price analyzer 280 may generate price change information by predicting price information for a specific period from the current time.

For example, the price analyzing unit 280 analyzes the event information corresponding to the event after the current point of time when the user searches for clothing, among the event information included in the recommended offline store information (for example, And the future price information can be predicted based on the event information analysis result and the current price (i.e., the current price included in the recommended offline store information in FIG. 8A). For example, based on the current time point TODAY shown in FIG. 8B, the second price line PG2 can be generated based on the event information of the second recommended offline store SEL2 corresponding to the future time point.

In one embodiment, the price analysis unit 280 may classify the price information of the recommended garment and display it on a map.

For example, the price analyzing unit 280 may calculate the price of the first pricing line PG1 based on the first pricing line PG1 (or the highest price at a specific point in time) and the second price line PG2 (or the lowest price at a specific point in time) The price of the recommended garment can be divided into L (i.e., an integer of 2 or more of L, for example, 3). In this case, the recommendation unit 250 may display the recommended offline store so as to be distinguished from other recommended offline stores, corresponding to the price grade of the recommended apparel sold in the recommended offline store.

For example, referring to FIG. 8D, the recommended garment sold in the second recommended in-store (SHOP2) may have a lowest price class (or first grade) Suggested offline stores (SHOP2) can be displayed in blue. Likewise, the recommended garment sold in the third recommended offline store SHOP3 may have a middle price rating (or second rating), and accordingly, the recommendation unit 250 may set the third recommended offline store SHOP3 It can be displayed in yellow. In addition, the recommended garment sold in the first recommended offline store SHOP1 may have the highest price grade (or third grade), and accordingly, the recommendation unit 250 may set the first recommended offline store SHOP1 to red It can be displayed in color. Therefore, the user can intuitively confirm not only the position of the recommended offline store but also the price of the recommended offline store.

As described with reference to FIGS. 7 through 8D, the recommendation unit 250 (or the garment recommendation system 100) may determine alternative and / or supplementary goods based on the user and / or the goods. In addition, the recommendation unit 250 (or the garment recommendation system 100) can generate price change information and price rating information of the recommended garment, and display the store information on the map. Accordingly, the user can select more user-customized products, can intuitively determine the purchase time of the product, the offline purchase location of the product, and activate the offline store.

9 is a flowchart showing a recommended garment providing method performed in the recommended garment system of Fig.

Referring to FIGS. 1 to 9, the clothing recommendation server 120 may request the seller server 130 to provide clothing information and offline store information (S901). The garment information is substantially the same as the garment information described with reference to Fig. 4A, so that redundant description will not be repeated. On the other hand, the clothing information may include offline store information (e.g., location information, inventory information, price information, etc.).

Seller server 130 may provide clothing information to apparel recommendation server 120 in response to the apparel information providing request (S902).

The clothing recommendation server 120 may calculate the clothing activity index for each garment based on the clothing information provided by the seller server 130 or the clothing information stored in the database 270 (S903).

Meanwhile, the user terminal 110 can access the apparel recommendation server 120 through a separate application or the like, or log in (S911).

In this case, the apparel recommendation server 120 requests the user terminal 110 to provide the user information (S912), and the user terminal 110 transmits the user information to the recommendation server 120 . Here, the user information is substantially the same as the user information described with reference to FIG. 4B, and includes user basic information (for example, sex, age and the like) and user addition information (for example, user location information and user activity information) .

The garment recommendation server 120 may calculate the user's preference based on the user information (or determine the preference goods based on the user group and the user group to which the user belongs), and calculate the user activity index (S914).

Meanwhile, the user terminal 110 may request the recommendation server 120 to search / recommend a specific apparel (S915).

In this case, the apparel recommendation server 120 may determine a recommended product (or recommended apparel) based on the user preference and the user activity index (S916). Also, as described with reference to FIGS. 8A to 8D, the apparel recommendation server 120 can generate a recommended product list including a recommended offline store, price change information, event information, and the like associated with the recommended product.

Then, the clothing recommendation server 120 can provide a recommendation product list to the user terminal 110 (S917).

Meanwhile, in FIG. 9, the clothing recommendation server 120 has provided clothing information provision request and clothing information provision among the seller servers 130, but the present invention is not limited thereto. For example, the clothing recommendation server 120 may calculate the clothing activity index based on the clothing information stored in the database 270 (or a separate server).

9, after the connection to the clothing recommendation server 120, the clothing recommendation server 120 calculates the user preference and the user activity index, and then determines a recommendation product. However, (120) is not limited thereto. For example, the garment recommendation server 120 determines a recommendation item based on the user activity index previously stored in the database 270, separately from the user information providing request, and transmits the new user information, The user preference and the user activity index may be re-calculated at a specific time, and the pre-stored data may be updated.

Although the clothes recommendation system according to the embodiments of the present invention has been described with reference to the drawings, the above description is illustrative, and it should be understood by those skilled in the art without departing from the technical idea of the present invention. Modifications and changes may be made.

100: Clothing recommendation system
110: User terminal
120: Apparel recommendation server
130: Merchant server
210:
220: collecting section
230: Affinity Analyzer
240: activity index calculating section
250: Input / output unit
260: Reference section
270: Database
280: Price Analysis Department
290:
310:
320: Purchase history analyzing unit
330: Preferred goods determining section
510: Speed calculation unit
520: speed correction unit
530:
710: Alternative product recommendation section
711, 712: First and second recommendation units
720: Supplementary Product Recommendation Department
721, 722: third and fourth recommendation parts

Claims (7)

In a clothing recommendation system connected to a user terminal via a network,
A database for storing user information and apparel information for a user associated with the user terminal;
A collection unit for collecting clothing information from the database;
A preference calculating unit for calculating user preference based on the preference questionnaire result information and purchase history information included in the user information;
An activity index calculating unit for calculating an apparel activity index based on the apparel dimension information and the apparel material information included in the apparel information; And
And a recommendation unit for determining a recommendation garment based on the user's preference, the user activity index included in the user information, and the clothing activity index,
Wherein the activity index calculation unit calculates the activity index based on the following equation (1): " (1) ""
[Equation 1]
AI (activity index) = S_k / SR_k * W1_k + W2_k *? Mi * Ri
In this case, AI is a clothing activity index, k is a clothing identification number, S_k is a circumference length of the garment, SR_k is a reference circumference length of a clothing group including the garment, W1_k is a first weight of the garment, W2_k Mi is the activity index of the i-th material of the garment, and Ri is the ratio of the i-th material of the garment.
delete The apparatus of claim 1, wherein the collecting unit collects activity information of a user from the user terminal, the activity information includes user position information of the user by time,
Wherein the activity index calculating unit calculates the user activity index of the user based on the user location information. The system of claim 1, wherein the user activity index is calculated based on the user location index.
4. The apparatus according to claim 3,
A speed calculating unit for calculating a moving speed according to time based on the user position information and generating a first moving speed graph showing a change in the moving speed according to a time change;
A speed correcting unit for generating a second moving speed graph by removing a section of the first moving speed graph exceeding a first reference speed; And
And a calculation unit for calculating the user activity index by integrating the second moving speed graph based on the area weight set for the actual area to which the position of the user belongs, and calculating the user activity index by considering the individual characteristics and preferences Customized on-line offline clothing recommendations system.
The apparatus according to claim 1,
Searching for a recommended offline store for selling the recommendation garment based on the offline store information included in the database,
Wherein the recommendation offline store is determined based on the current location information of the user terminal and the address information of the user. The system of claim 1, wherein the recommendation offline store is determined based on the current location information of the user terminal and the address information of the user.
6. The method of claim 5,
Generates price change information according to time based on price information of the recommended garment included in the database, and generates price change information by associating event information of the recommended offline store with a price change point at which the price information changes And a price analysis unit. The system includes a customer-specific on-line linked clothing recommendation system that takes individual characteristics and preferences into account using the Big Data.
7. The information processing apparatus according to claim 6, wherein the price analyzing unit classifies the price information of the recommendation garment,
And the recommendation unit displays the recommended offline store together with the graded price information on a map. The on-line linked clothing recommendation system is customized based on individual characteristics and preferences using the big data.

Images