KR102582813B1 - Method and apparatus for providing personalized recommendation service for offline purchase - Google Patents

Abstract

The present invention is a method of providing a personalized recommendation service for offline purchases, where a user visits an offline store, tries on coordinated clothes and purchases them, or finds or recommends matching clothes in the store and calculates a coordination score to see if they match the clothes they already own. By looking at and judging the suitability in advance, time-consuming processes such as fitting can be minimized and the offline purchasing experience can be improved, thereby improving consumer purchase satisfaction and increasing the revisit rate.

Description

Method and apparatus for providing personalized recommendation service for offline purchases {METHOD AND APPARATUS FOR PROVIDING PERSONALIZED RECOMMENDATION SERVICE FOR OFFLINE PURCHASE}

The present invention relates to a method and device for providing personalized recommendation services for offline purchases.

Currently, as interest in fashion increases, many people own a variety of clothing items, and the desire to purchase clothing items also increases, leading to the development or provision of various industries or services related to online shopping or offline shopping. People can satisfy their purchasing desires by selectively purchasing desired clothing items online or offline depending on their situation or preference.

However, in the case of conventional online shopping, even if the buyer carefully decides to purchase by referring to information provided online, such as product image information, seller's product description, and buyers' product reviews, the product that is actually received There were frequent cases of purchase failure due to significant differences from the information provided online. In addition, the return and exchange process due to purchase failure also has the problem of being cumbersome and consuming time and effort.

Meanwhile, in the case of conventional offline shopping, unlike online shopping, buyers purchase products after experiencing them in person at an offline store, so the difference between the buyer's expectations and actual experience is not large, so satisfaction with shopping is high. However, because offline purchases consume a lot of time, there is a problem in that there are physical and time constraints on the buyer. Specifically, buyers must spend a lot of time to purchase products, such as the time they spend traveling to an offline store and the time they browse products to select the product they want, and even then, there are cases where they cannot find or buy what they want. It occurred frequently. This has the problem of making the buyer feel like they wasted time, causing them to remember their experience with the offline store negatively and feel reluctant to visit again.

Therefore, a method is needed to enable users to shop efficiently by taking advantage of only the advantages of online and offline shopping while resolving the problems of online and offline shopping.

Republic of Korea Patent Publication No. 10-2016-0064379, 2016.06.08.

The purpose of the present invention to solve the problems described above is to provide a method and device for providing personalized recommendation services for offline purchases.

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

A method of providing a personalized recommendation service for offline purchases according to one aspect of the present invention to solve the above-described problem includes obtaining clothing information related to clothing owned by a user from a user terminal, based on the clothing information held by the user. Extracting usage information for the clothing held, extracting a product that needs to be purchased from the clothing held based on the usage information, searching at least one first offline store for the product that needs to be purchased, and 1. Providing search results to the user terminal, wherein the usage information includes the number of times the clothing has been worn and the date of purchase.

In the present invention, the first offline store search step is to sort the at least one first offline store in the order of the closest distance to the user, based on the distance from the user's current location, and to select the product that needs to be purchased. During extraction, when a plurality of products needing to be purchased are extracted, the first offline store search step selects the at least one first offline store in the order of the closest distance to the user, based on the distance from the user's current location. Sorting according to the inventory information for each of the plurality of products needing to be purchased, the at least one first offline store sorted in order of closest distance to the user is in stock among the plurality of products needing to be purchased You can rearrange them in order of the number of products you need to purchase.

In the present invention, the step of providing display information of the product to be purchased at a first offline store selected by the user terminal among the searched first offline stores to the user terminal may be further included.

In the present invention, the method may further include receiving, from the user terminal, a scheduled visit time for a first offline store selected by the user terminal among the searched first offline stores.

In the present invention, based on the information on the clothing held, analyzing characteristic information on the held clothing, specifying the user in a multidimensional space indicating the degree of similarity between users based on the analysis result, the multidimensional space Recommending a product (hereinafter referred to as a recommended product) that matches the product that needs to be purchased, based on clothing information held in a user group located in a space adjacent to the user, searching at least one second offline store for the recommended product. The method may further include providing a second search result to the user terminal.

In the present invention, the characteristic information analysis step includes extracting at least one clothing factor representing an attribute of clothing from the held clothing information, and extracting characteristic information about the held clothing information based on the at least one clothing factor. Analyzing and calculating the characteristic value of the user, the user specification step includes obtaining coordinate information corresponding to the characteristic value in the multidimensional space, and specifying the user based on the obtained coordinate information. It may include steps.

In the present invention, the second offline store search step sorts the at least one second offline store in the order of the closest distance to the first offline store, based on the distance from the first offline store, When recommending a recommended product, if a plurality of recommended products are recommended, the first offline store search step selects the at least one second offline store from the first offline store based on the distance from the first offline store. The plurality of second offline stores are sorted in order of short distance, and are sorted in order of short distance to the first offline store based on inventory information for each of the plurality of recommended products. Among the recommended products, they can be rearranged in order of the number of recommended products in stock.

In the present invention, the step of determining whether there is clothing similar to the recommended product among the clothing held by the user, and if it is determined that there is clothing held similar to the recommended product, providing a coordination score between the similar clothing held and the product that needs to be purchased. Additional steps may be included.

In the present invention, it may further include obtaining image information of a product desired to be purchased from the user terminal, and providing coordination recommendations to the user terminal based on the clothing information and image information of the product desired to be purchased. You can.

An apparatus according to another aspect of the present invention for solving the above-described problem includes a memory that stores one or more instructions, and a processor that executes the one or more instructions stored in the memory, wherein the processor executes the one or more instructions. Obtaining clothing information related to the clothing owned by the user from the user terminal by executing, extracting usage information about the clothing held based on the clothing information, and selecting among the clothing held based on the usage information. The steps of extracting a product that needs to be purchased, searching at least one first offline store for the product that needs to be purchased, and providing a first search result to the user terminal are performed.

In addition to this, another method for implementing the present invention, another device, another system, and a computer-readable recording medium recording a computer program for executing the method may be further provided.

According to the present invention as described above, based on the clothing owned by the user, clothing that needs to be repurchased (goods to be purchased) is extracted from the clothing owned, and offline stores where the clothing can be purchased are presented, allowing the user to You can quickly purchase fashion items offline.

Additionally, by accurately analyzing the user's characteristics (i.e., tastes, preferences, etc.) based on the clothing the user owns, a personalized coordination that takes the user's characteristics into consideration can be recommended.

In addition, by collecting and analyzing clothing information owned by multiple users, the characteristics of each user (i.e., tastes, preferences, etc.) are determined, and through this, users with the same or similar characteristics are grouped together to determine the preferred coordination of users within each group. Patterns can be identified. In addition, since users in each group have the same or similar characteristics, fashion items that match the product the user needs to purchase are extracted using the preferred coordination of users in the same group or adjacent groups, and these are extracted as recommended products. By presenting it to the user, personalized coordination recommendations can be made.

In addition, by suggesting offline stores where recommended products can be purchased in conjunction with offline stores where products that need to be purchased can be purchased, users can quickly purchase recommended products along with products that need to be purchased, enabling efficient offline shopping.

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 below.

1 is a diagram schematically showing a system for providing a personalized recommendation service for offline purchases according to the present invention.
Figure 2 is a flowchart of a method for providing a personalized recommendation service for offline purchase of a product requiring purchase according to the present invention.
Figure 3 is a flowchart of a method for providing a personalized recommendation service for offline purchases of recommended products according to the present invention.
Figure 4 is a flowchart of a specific method of step S210 of Figure 3.
FIG. 5 is a flowchart of a specific method of step S220 of FIG. 3.
Figure 6 is a diagram illustrating an example of a multidimensional space representing the degree of similarity between users according to the present invention and a process for creating a user group by grouping users according to their degree of similarity within the space.
Figure 7 is a diagram schematically showing the configuration of a device that performs a method of providing a personalized recommendation service for offline purchases according to the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to provide a general understanding of the technical field to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for describing embodiments and is not intended to limit the invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. As used in the specification, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other elements in addition to the mentioned elements. Like reference numerals refer to like elements throughout the specification, and “and/or” includes each and every combination of one or more of the referenced elements. Although “first”, “second”, etc. are used to describe various components, these components are of course not limited by these terms. These terms are merely used to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may also be a second component within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless clearly specifically defined.

The present invention provides the value of direct experience through offline shopping in order to solve the problems of conventional online shopping or offline shopping, while resolving the pain point of offline shopping that a lot of time is consumed to purchase a product. By collecting and analyzing various information about the buyer's purchasing experience, it is possible to present and recommend products that the buyer needs, as well as information on offline stores where the product can be purchased. Accordingly, through offline shopping, buyers gain the value of a direct experience that cannot be experienced through online shopping, and are provided with products that suit their needs or tastes and information on offline stores where they can purchase the products. It allows you to shop quickly and efficiently even offline.

Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

1 is a diagram schematically showing a system for providing a personalized recommendation service for offline purchases according to the present invention.

Figure 2 is a flowchart of a method for providing a personalized recommendation service for offline purchase of a product requiring purchase according to the present invention.

Figure 3 is a flowchart of a method for providing a personalized recommendation service for offline purchases of recommended products according to the present invention.

Figure 4 is a flowchart of a specific method of step S210 of Figure 3.

FIG. 5 is a flowchart of a specific method of step S220 of FIG. 3.

Figure 6 is a diagram illustrating an example of a multidimensional space representing the degree of similarity between users according to the present invention and a process for creating a user group by grouping users according to their degree of similarity within the space.

Figure 7 is a diagram schematically showing the configuration of a device that performs a method of providing a personalized recommendation service for offline purchases according to the present invention.

Referring to Figure 1, the system 1 (hereinafter referred to as system) for providing a personalized recommendation service for offline purchases according to an embodiment of the present invention includes a server 10, a user terminal 20, and an offline store terminal ( 30) and a communication network 40.

In the system 1 according to the present invention, the subject of the invention may be a computing device or a server device, and specifically, the server 10 may be the subject of execution, but is not limited thereto, and in some cases, the user terminal 20 Alternatively, the offline store terminal 30 may be the entity that performs the system 1.

The server 10 is a computing device for providing a method of providing a personalized recommendation service for offline purchases according to an embodiment of the present invention, and provides personalized recommendations to users based on the clothing they own through the user terminal 20. It can perform various functions to provide services. For convenience of explanation, FIG. 1 illustrates using one server 10, but the system 1 according to an embodiment of the present invention may be configured to include one or more servers. Additionally, the server 10 may be configured as a cloud server.

The user terminal 20 is a terminal device of a user (i.e., a client) who uses a personalized recommendation service based on clothing owned by the user provided through the server 10, such as a smart phone, tablet PC, desktop, laptop, smart TV. It may be various computing devices such as the like.

The user terminal 20 may include an application that provides a personalized recommendation service based on clothing owned by the user. Accordingly, the user can be provided with a personalized recommendation service based on the clothing owned by the user through the application of the user terminal 20.

Here, the user may refer to a buyer who receives the personalized recommendation service and purchases a product at an offline store.

The offline store terminal 30 is a terminal device of an offline store linked to the personalized recommendation service provided by the server 10 to the user. For example, various computing devices such as smart phones, tablet PCs, desktops, laptops, smart TVs, etc. It can be.

The offline store terminal 30 can provide the server 10 with various information related to the offline store, such as store location, types of products in stock, and inventory information on the products in stock.

The server 10 may recommend the optimal offline store where the user can purchase the recommended product based on the type and inventory information of products held by each connected offline store.

The server 10 can update inventory information on products held by the offline store 30 in real time. For example, the server 10 may receive images and inventory information of products held through an inventory input application provided to employees of a linked offline store. Through this, small merchants can upload information such as inventory information to the online server 10 using a simple inventory input service (i.e. inventory input application) in offline stores without investing in new infrastructure. Additionally, for example, the server can obtain inventory information in real time by linking with the inventory management program of each connected offline store.

The communication network 40 enables transmission and reception of various information between the server 10, the user terminal 20, and the offline store terminal 30. The communication network 40 may use various types of communication networks, for example, WLAN (Wireless LAN), Wi-Fi, Wibro, Wimax, HSDPA (High Speed Downlink Packet Access), etc. Wireless communication methods or wired communication methods such as Ethernet, xDSL (ADSL, VDSL), HFC (Hybrid Fiber Coax), FTTC (Fiber to The Curb), and FTTH (Fiber To The Home) may be used.

However, the communication network 40 is not limited to the communication methods presented above, and may include all types of communication methods that are well known or will be developed in the future in addition to the communication methods described above.

Hereinafter, a specific process for providing a method of providing a personalized recommendation service for offline purchases will be described using the server 10, user terminal 20, and offline store terminal 30 included in the system 1. .

Referring to FIG. 2, the method of providing a personalized recommendation service for offline purchases performed by the server 10 according to an embodiment of the present invention includes the step of obtaining clothing information related to the user's clothing from the user terminal ( S110), extracting usage information about the owned clothing based on the held clothing information (S120), extracting a product that needs to be purchased from the owned clothing based on the used information (S130), the purchased product It may include searching for at least one first offline store (S140) and providing a first search result to the user terminal (S150). Below, a detailed description of each step is described.

The server 10 may obtain clothing information related to the user's clothing from the user terminal 20 (S110).

Here, the owned clothing refers to clothing items that the user already owns, and is used to encompass all of the user's own clothing and various clothing items related to clothing. For example, the user's clothing may include clothing such as tops, bottoms, outerwear, and underwear, as well as other shoes, bags, hats, and accessories.

In one embodiment, the user terminal 20 may directly receive clothing information related to the clothing owned by the user from the user. Here, the information on the owned clothing may include the type, color, shape and shape (image), material, fabric pattern, price, size, number of wears, purchase date, etc. of the owned clothing. In this case, the server 10 may obtain clothing information directly input from the user through the user terminal 20.

At this time, among the clothing information held, the type, color, shape and shape (image), material, fabric pattern, price, size, purchase date, etc. of the clothing held are fixed information, and the initially entered values can be maintained as is. On the other hand, among the clothing information held, the number of times worn is variable information, and the initially entered value is a default value (for example, 0), and may be updated each time the user wears the clothing.

For example, when the user wears a specific clothing item among the clothing items, if the user selects the clothing item worn along with the date of wearing through the application included in the user terminal 20, the number of times the clothing item has been worn is 1 time. It can be increased and updated.

In another embodiment, the user terminal 20 may receive information on the user's clothing as an image (photo, video, etc.) from the user. For example, a user can input a photo or video of clothing, such as clothes, shoes, or bags, taken using a camera into the user terminal 20. In this case, the server 10 obtains an image containing information on the user's clothing from the user terminal 20, and information about the clothing (e.g., type, color, shape, and shape of the clothing held) from the acquired image. By recognizing (image), material, fabric pattern, size, etc.), information on the clothing owned by the user can be obtained. For example, the server 10 uses an object detection algorithm based on deep learning (e.g., CNN; convolution neural network) to detect the type, color, and type of clothing held in the image acquired from the user terminal 20. Information such as shape and shape (image), material, fabric pattern, and size can be recognized, and the recognized information can be obtained as information on the clothing owned by the user. At this time, the purchase date among the owned clothing information may be set as the default date on which the owned clothing information was entered, and may then be modified by the user to the actual purchase date.

In another embodiment, the user terminal 20 recognizes clothing product code information (e.g., barcode, QR code, etc.) provided for clothing, and provides manufacturer and product information (type, color, and type of clothing held) from the recognized clothing product code information. , shape and shape (image), material, fabric pattern, price, size, etc.) can be obtained. Alternatively, the user terminal 20 may recognize clothing product code information provided for clothing and transmit the recognized clothing product code information to the server 10.

At this time, the server 10 may establish a database that stores and manages clothing product information. In this case, when the server 10 obtains clothing product code information from the user terminal 20, the server 10 obtains manufacturer and product information (type, color, shape, and shape of the clothing held) corresponding to the clothing product code information from the constructed database. By searching (image), material, fabric pattern, price, size, etc.), information on the user's clothing can be obtained. Additionally, when building a database, the server 10 can obtain clothing product information from clothing manufacturers or sellers and convert it into a database. At this time, the purchase date among the owned clothing information may be set as the default date on which the owned clothing information was entered, and may then be modified by the user to the actual purchase date.

In another embodiment, when a user purchases clothing or clothing-related items from a clothing seller such as an online shopping mall, the user terminal 20 obtains information about the clothing or clothing-related items purchased by the user from the clothing seller. , this can be transmitted to the server 10. For example, the user terminal 20 may obtain image information about clothing or clothing-related items purchased by the user from the clothing seller and transmit this to the server 10. Alternatively, the user terminal 20 acquires identification information about the clothing or clothing-related item purchased by the user (e.g., identification information of the clothing seller and identification information about the clothing product, etc.), and sends this identification information to the server (10). ) can be transmitted. At this time, the server 10 may obtain information on the clothing or clothing-related items from the clothing seller based on the identification information.

As described above, the server 10 can obtain clothing information about clothing owned by the user through the user terminal 20 in various ways. In this way, clothing information obtained from each user can be used as big data.

In one embodiment, the server 10 may build a big data pipeline to collect clothing information owned by each user, analyze, process, and use it according to the purpose. Additionally, the server 10 may use artificial intelligence technology, such as machine learning and deep learning, to analyze and utilize the user's clothing information collected from each user according to the purpose. At this time, the server 10 may perform learning and build a learning model using the user's clothing information collected from each user as learning data. In this way, when using the user's clothing information as big data or learning data, the system 1 according to an embodiment of the present invention may configure and use a separate server in addition to the server 10.

In this way, by using the clothing information obtained from each user as big data, the server 10 provides a customized service that recommends coordination that suits each situation, clothing owned, preference, taste, etc. to users who use the personalized recommendation service. can be provided.

The server 10 may extract usage information about the user's clothing based on the clothing information obtained in step S110 (S120).

Here, the usage information may include the number of times the clothing is worn and the date of purchase. In other words, the server 10 can use the number of wears and purchase date among the clothing information held as usage information.

According to the embodiment, the server 10 obtains clothing information from the user terminal 20 of users at regular intervals (e.g., daily, weekly, monthly, etc.), and based on this, the clothing owned by the user is acquired at regular intervals. By reflecting the information, the usage information for the user's clothing information can be updated.

The server 10 may extract products that need to be purchased from the owned clothing based on the usage information extracted in step S120 (S130).

Specifically, the server 10 stores clothing among the user's owned clothing, the number of times of wearing being more than a preset standard number, and the period between the purchase date and the current date (i.e., the period elapsed from the purchase date) being more than the preset reference period. can be extracted as a product that needs to be purchased.

For example, if the standard number of times is 100 and the standard period is 5 years, the number of times clothing A is worn is 20 times, the period elapsed from the date of purchase is 3 years, and the number of times clothing B is worn is 10 times, and the purchase date is If the elapsed period is 1 year, the number of times the owned clothing C has been worn is 120 times, and the standard period is 5 years, the server 10 can extract the owned clothing C as a product that needs to be purchased.

Depending on the embodiment, the server 10 may extract the owned clothing as a product that needs to be purchased if the number of wears is more than a preset maximum number even if the period of time elapsed from the date of purchase does not exceed the reference period. For example, if the user has worn clothing A less than 5 years ago but has worn it more than the maximum number of times, for example, 200 times, the server 10 may extract clothing A as a product that needs to be purchased. .

Depending on the embodiment, the server 10 may extract the owned clothing as a product that needs to be purchased if the period of time elapsed from the purchase date is more than a preset maximum period even if the number of wears does not exceed the standard number. For example, if the number of times the user has worn clothing A does not exceed 100, but it has been purchased for more than the maximum period of time, for example, 8 years, the server 10 may extract the clothing A as a product that needs to be purchased. You can.

In this way, the server 10 can provide a personalized recommendation service to the user by extracting clothing that is determined to require repurchase based on the number of wears or purchase dates from among the clothing owned by the user as products that need to be purchased and presenting them to the user.

Meanwhile, the server 10 may extract products that need to be discarded from the owned clothing based on usage information and present them to the user.

Specifically, if the period elapsed from the purchase date is more than the maximum period and the number of wears is less than the preset minimum number of times, the server 10 may extract the owned clothing as a product that needs to be discarded and present it to the user. For example, if the user has purchased clothing A for a maximum period of time, for example, more than 8 years, but the number of times it has been worn has not exceeded the minimum number of times, for example, 10, the server 10 discards the clothing A. It can be extracted into required products. Accordingly, the user can dispose of clothing that does not suit the user's taste (is not worn frequently).

Although it has been described above that the server 10 extracts one product that needs to be purchased and a product that needs to be discarded, the server 10 is not limited to this and a plurality of products that need to be purchased and products that need to be discarded may be extracted.

The server 10 may search for at least one first offline store for the product needed to be purchased extracted in step S130 (S140).

Depending on the embodiment, the server 10 may sort the at least one first offline store in the order of the closest distance to the user, based on the distance from the user's current location.

At this time, the server 10 may extract the offline store that holds the product that needs to be purchased from among the numerous offline stores linked to the personalized recommendation service as the first offline store. That is, the server 10 does not extract offline stores that do not have the products required to be purchased as the first offline store.

The first offline stores extracted in this way are sorted and searched in order of distance from the user's location.

For example, the first offline stores A, B, and C all have the product A that needs to be purchased. The first offline store A is 1 km away from the user's location, and the first offline store B is located at the user's location. If the distance to the first offline store C is 5km and the distance from the user's location is 0.8km, the server 10 orders the calculated distances in the order of the shortest, that is, the first offline store C, A, and B. You can sort by .

According to an embodiment, when the server 10 extracts a plurality of products that need to be purchased in step S130, the server 10 connects the at least one first offline store to the user based on the distance from the current location of the user. The plurality of first offline stores are sorted in the order of the closest distance, and are sorted in the order of the closest distance to the user, based on inventory information for each of the plurality of products that need to be purchased. Among the products that need to be purchased, you can rearrange them in order of the number of products that are in stock.

At this time, the server 10 may extract, as the first offline store, an offline store that holds at least one product that needs to be purchased among the plurality of products that need to be purchased from among numerous offline stores linked to the personalized recommendation service. That is, the server 10 does not extract an offline store that does not have any of the plurality of products that need to be purchased as the first offline store.

The first offline stores extracted in this way are sorted and searched in order of distance from the user's location. Specifically, scores may be given in the order of the calculated distance to the user's location, and may be sorted in order of higher scores.

For example, the first offline stores A, B, and C all have at least one of the plurality of products that need to be purchased, where the first offline store A is 1 km away from the user's location, and the first offline store B is If the distance from the user's location is 5km and the distance from the user's location to the first offline store C is 0.8km, the server 10 gives 3 points to the first offline store C with the shortest calculated distance. , 2 points are given to the first offline store A, which has the second shortest calculated distance, and 1 point is given to the first offline store B, which has the longest calculated distance, so that the first offline stores C and A with the highest given score , can be sorted in the order of B.

And the three first offline stores, sorted based on distance, are rearranged in the order of the number of products in stock among the plurality of products to be purchased. Specifically, the scores given to each of the first offline stores A, B, and C are weighted according to the number of products in stock that need to be purchased to calculate the final score, and the calculated final score can be rearranged in descending order. there is.

In other words, if 5 products that need to be purchased (A, B, C, D, E) are extracted, the first offline store A has inventory for products that need to be purchased A, B, C, D, and E among the 5 products that need to be purchased. The first offline store B has inventory for products A, B, and C out of the five products that need to be purchased, and the first offline store C only has inventory for product A that needs to be purchased. If so, the number of products that need to be purchased in stock at each of the first offline stores A, B, and C is 5, 3, and 1, so weights may be assigned based on this.

For example, the first offline store A is 1 km away from the user's location, the first offline store B is 5 km away from the user's location, and the first offline store C is 0.8 km away from the user's location. According to km, the first offline stores C, A, and B are sorted in the order of The number of products that need to be purchased is 5, the first offline store B has 3 products in stock that need to be purchased, and the first offline store C has 1 product that needs to be purchased in stock, so A weight of 3 points is given to the first offline store A with the largest number of products in stock that need to be purchased, and a weight of 2 points is given to the first offline store B that has the second largest number of products in stock that need to be purchased. A weight of 1 point may be given to the first offline store C, which has the smallest number of products in stock that need to be purchased. Accordingly, since the calculated final scores of the first offline stores A, B, and C are 6 points, 2 points, and 3 points, respectively, they can be rearranged from the order of C, A, and B to the order of A, C, and B. .

At this time, for the first offline stores with the same calculated final score, the first offline store with a shorter distance from the user's location may be determined as priority.

Depending on the embodiment, the server 10 may further consider store business hours when searching for the first offline store. Specifically, the first offline stores may be sorted in order of the most remaining business hours based on the current time.

The server 10 may provide the search result (first search result) in step S130 to the user terminal 20 (S150).

The user can check the product that needs to be purchased and the first offline store where the product that needs to be purchased can be purchased through the user terminal 20, and visit a store located close to the user's location to purchase the product. Through this, users can shop offline more efficiently because they can directly experience value through offline shopping and purchase products directly without having to search separately through recommendations for products they need to purchase.

Meanwhile, the present invention may further include providing, to the user terminal 20, display information of the product to be purchased at a first offline store selected by the user terminal 20 among the searched first offline stores. .

Specifically, the server 10 receives a specific first offline store selected from the user terminal 20 among the searched first offline stores, and sends the location information where the product to be purchased is displayed at the selected specific first offline store to the user terminal 20. ), so that the user can visit the first offline store and immediately find the product that needs to be purchased.

At this time, the server 10 may receive the display information in advance from the first offline store terminal 30.

In addition, the present invention may further include receiving, from the user terminal 20, a scheduled visit time for the first offline store selected by the user terminal 20 among the searched first offline stores.

Specifically, a user who has been recommended a first offline store for offline purchase of a product that needs to be purchased can set a scheduled visit time for the first offline store and immediately receive the product to be purchased when visiting the store. For example, the first offline store linked to the personalized recommendation service of the present invention may package the product in advance according to the user's visit reservation and then deliver the product through a drive-through method. To this end, users can make payment for products that need to be purchased in advance on the service platform.

Depending on the embodiment, the server 10 may provide a stock alarm service to the user. If the product to be purchased is not in stock at the offline store desired by the user, other than the first offline store searched by the server 10, the user can apply for an inventory alarm service for the product to be purchased. Thereafter, when a product requiring purchase is received at the relevant offline store, a stock alarm is provided in the form of a message to the user terminal 20. At this time, the user enters the expected visit time for the relevant offline store through the user terminal 20. If you do this, you will be able to immediately receive the products you plan to purchase when you visit the store.

Referring to FIG. 3, the method of providing a personalized recommendation service for offline purchases performed by the server 10 according to an embodiment of the present invention analyzes characteristic information about the owned clothing based on the held clothing information. A step (S210), a step of specifying the user in a multidimensional space indicating the degree of similarity between users based on the analysis results (S220), clothing information held in a user group located in a space adjacent to the user in the multidimensional space Based on this, a step of recommending a product (hereinafter referred to as a recommended product) matching the product to be purchased (S230), a step of searching at least one second offline store for the recommended product (S240), and the second search result. A step of providing information to a user terminal (S250) may be further included. Below, a detailed description of each step is described.

The server 10 may analyze characteristic information about the owned clothing based on the user's clothing information obtained in step S110 (S210).

Specifically, referring to FIG. 4, the server 10 extracts at least one clothing factor representing the attributes of the clothing from the held clothing information (S212), and adds to the held clothing information based on the at least one clothing factor. By analyzing the characteristic information about the user, the characteristic value of the user can be calculated (S214).

Here, the clothing factor may mean information representing the attributes of clothing or clothing items related to clothing. For example, the type, color, shape and shape (image), material, fabric pattern, price, size, brand name, etc. of the clothing owned can be used as clothing factors.

In one embodiment, the server 10 may extract at least one clothing factor representing clothing attributes from clothing information held by the user. Additionally, the server 10 may analyze characteristic information about the user's clothing information based on at least one extracted clothing factor and calculate the user's characteristic value.

For example, the server 10 extracts the type, color, shape and shape (image), and material of clothing from clothing A held by the user as a clothing factor, and extracts the type, color, and When the form and shape (image), price, and brand name are extracted as clothing factors, the information about the clothing owned by the user is based on 4 clothing factors for clothing A and 5 clothing factors for clothing B. By analyzing characteristic information, characteristic values can be calculated.

Here, the characteristic information is information that can be determined by analyzing the entire clothing owned by each user, and includes preference information indicating taste information for clothing, preferred style for clothing, and what types of clothing or clothing items the user likes. It may be information that can represent, etc.

Additionally, the characteristic value may be expressed as a single value representing characteristic information about the user's clothing, or it may be expressed as a multidimensional value in the form of a vector or matrix rather than a single value. If the characteristic value is expressed as a multidimensional value based on the clothing factor, the multidimensional characteristic values for each user can be gathered and placed in a high-order multidimensional space. Multidimensional space will be described later.

In other words, the server 10 can determine each user's tastes and preferred styles by obtaining clothing information owned by each user and analyzing the characteristic information.

In the present invention, when analyzing characteristic information about clothing information held by a user according to an embodiment, the server 100 may use social data.

The server 10 may obtain social data through the user terminal 20. At this time, social data may include various information that can be obtained from social media (eg, social network services (SNS), blogs, Twitter, etc.) and users.

For example, posts posted by users through social media, comments on other people's posts, reaction information (e.g., like click information, etc.), etc. can be obtained as social data. In addition, posts posted by other users related to the user (e.g., the user's acquaintances) through social media, comments on other people's posts, and reaction information (e.g., like click information, etc.) can be obtained as social data. You can. Additionally, various information related to clothing and clothing items posted through social media can be obtained as social data. In addition, weather information, etc., which is not directly related to clothing but may indirectly affect it, can be obtained through social data.

Additionally, the social data may include star rating feedback data for the results of each individual user's own coordination or satisfaction rating data for the recommended coordination. In one embodiment, the server 10 may acquire social data associated with the user and analyze characteristic information about the user's clothing information by reflecting the acquired social data associated with the user.

That is, the server 10 may data mine the social data together with the user's clothing information obtained in step S110 to analyze characteristic information about the user's clothing information.

In this way, by using not only the user's own clothing information but also social data linked to the user, the user's tastes can be analyzed more precisely. Additionally, by mining various social data, user tastes that reflect the latest trends can be considered.

The server 10 may specify the user in a multidimensional space indicating the degree of similarity between users based on the results analyzed in step S210 (S220).

Specifically, referring to FIG. 5, the server 10 may obtain coordinate information corresponding to the characteristic value in the multidimensional space (S222) and specify the user based on the obtained coordinate information (S224).

Here, the multidimensional space is a space for indicating the degree of similarity between users, and refers to a multidimensional coordinate space that can be represented by mapping users with similar characteristic information to the same or similar space based on the characteristic information of each user. You can.

In one embodiment, the server 10 may obtain coordinate information corresponding to the user's characteristic information (i.e., characteristic value) in a multidimensional space and map the user to the obtained coordinate information.

Referring to FIG. 6, the server 10 may map each user to a specific location in the multidimensional space 300 and 400 using characteristic information analyzed based on clothing information owned by each user. Each point shown on the multidimensional space 300 and 400 of FIG. 6 represents each user mapped to a specific location on the multidimensional space 300 and 400.

At this time, the server 10 creates at least one user group 310, 410 based on the degree of similarity (i.e. characteristic information such as taste, preferred style, etc.) between each user mapped to a specific location in the multidimensional space 300, 400. can be created.

In one embodiment, the server 10 may analyze characteristic information for each of at least one user based on clothing information for at least one user obtained from at least one user terminal 20. The server 10 may calculate coordinate information in the multidimensional space 300 and 400 corresponding to characteristic information for each of at least one user, and map each user to the calculated coordinate information. The server 10 generates at least one user group 310, 410 by grouping users showing similar characteristic information based on characteristic information of each at least one user mapped on the multidimensional space 300, 400. can do.

That is, users within the same user group 310 and 410 may be understood to have the same or similar characteristic information (eg, taste, preferred style, etc.). For example, users within the same user group 310 and 410 may own the same or similar clothing or clothing items.

In creating user groups in a multidimensional space, the server 10 applies artificial intelligence (i.e., data learning-based machine learning, deep learning, etc.) technologies such as clustering and collaborative filtering to create user groups that are similar to each other. Users with characteristics can be grouped.

Additionally, when creating a user group in a multidimensional space, the server 10 may map users on the multidimensional space 300 in a parametric model-based form, or in a graph-based form. Users can also be mapped on the multidimensional space 400 in a form.

The parametric model-based form or graph-based form is only mentioned as an example, and embodiments of the present invention are not limited to the above-exemplified form and may be mapped or configured in various forms.

As shown in FIG. 6, the server 10 may pre-construct a multidimensional space in which users with the same or similar characteristic information are grouped and classified among a plurality of users. Afterwards, when the server 10 obtains clothing information from a new user or new clothing information from an existing user, it analyzes the characteristic information and maps the user to a specific location in a pre-built multidimensional space. , Based on the mapped location, the user can be grouped into user groups with the same or similar characteristic information in a multidimensional space.

In one embodiment, when the server 10 obtains new clothing information from the user terminal 20 of an existing user, the server 10 reflects the new clothing information along with the existing clothing information to determine the characteristics of the user's clothing information. Information can be updated. In this case, the server 10 may calculate coordinate information corresponding to the updated characteristic information in multidimensional space based on the updated characteristic information and change the location of the user using the calculated coordinate information. At this time, as the user's location changes in the multidimensional space, the user may be grouped into a different user group than the existing user group. In this case, it can be inferred that the user's characteristics (i.e., taste or preferred style, etc.) have changed.

In addition, the server 10 acquires information on clothing held by users from the user terminal 20 at regular intervals (e.g., daily, weekly, monthly, etc.), and based on this, reflects the information on clothing held by users acquired at regular intervals. Thus, characteristic information about the user's clothing information can be updated. Additionally, the server 10 may change the user's location in the multidimensional space based on characteristic information updated at regular intervals to periodically reflect changes in fashion taste.

In step S220, the server 10 selects a product (hereinafter referred to as recommendation) that matches the product to be purchased based on clothing information held in a user group located in a space adjacent to the user in the multidimensional space based on a specific result in the multidimensional space. product) can be recommended (S230).

In one embodiment, the server 10 detects a user group located in a space adjacent to the user among at least one user group created in a multidimensional space, and based on information on clothing held by users in the detected user group, the purchase necessity is determined. Clothing similar to the product can be extracted, and clothing that users in the detected user group have coordinated with the extracted similar clothing can be obtained. Additionally, the server 10 may select the obtained clothing as the recommended product and recommend it to the user.

In another embodiment, the server 10 detects a user group located in a space adjacent to the user among at least one user group created in a multidimensional space, and based on information on clothing held by users in the detected user group, the purchase necessity is determined. Extract clothing similar to the product, and based on star rating feedback data or satisfaction rating data for the coordination matched with the extracted similar clothing given by users in the detected user group, similar clothing extracted from the coordination with a high star rating. You can obtain clothing coordinated with . Additionally, the server 10 may select the obtained clothing as the recommended product and recommend it to the user.

At this time, the detected user group may be the same user group as the user, or may be an adjacent user group adjacent to the user's group.

Each user group created in a multidimensional space is a group composed of users with the same or similar characteristic information. Accordingly, users within the same user group or adjacent user groups have a high probability of owning the same or similar clothing or clothing items. Accordingly, the server 10 may increase the probability of recommending clothing (coordination) that matches the product the user currently needs to purchase by using clothing owned by other users located adjacent to the user in a multidimensional space.

Although it has been described above that the server 10 recommends one recommended product, the present invention is not limited to this and a plurality of recommended products may be recommended.

The server 10 may search for at least one second offline store for the recommended product recommended in step S230 (S240).

Depending on the embodiment, the server 10 may sort the at least one second offline store in the order of the closest distance to the first offline store, based on the distance from the first offline store.

At this time, the server 10 may extract the offline store holding the recommended product as the second offline store from among numerous offline stores linked to the personalized recommendation service. That is, the server 10 does not extract offline stores that do not have the recommended products as second offline stores.

The second offline stores extracted in this way are sorted and searched in the order of their distance from the location of the selected first offline store.

For example, the second offline stores A, B, and C all have the recommended product A. The second offline store A is 0.5 km away from the location of the first offline store, and the second offline store B is 0.5km away from the location of the first offline store. If the distance from the location of the first offline store is 2km, and the distance of the second offline store C from the location of the first offline store is 5km, the server 10 stores the second offline store in order of the shortest calculated distance, that is, Stores can be sorted in the following order: A, B, and C.

According to an embodiment, when the server 10 recommends a plurality of recommended products in step S230, the server 10 selects the at least one second offline store as the first offline store based on the distance from the first offline store. The at least one second offline store sorted in the order of shortest distance to the offline store, and based on inventory information for each of the plurality of recommended products, sorted in the order of shortest distance to the first offline store. The can be rearranged according to the order of the number of recommended products in stock among the plurality of recommended products.

At this time, the server 10 may extract an offline store holding at least one recommended product among the plurality of recommended products as a second offline store from among numerous offline stores linked to the personalized recommendation service. That is, the server 10 does not extract an offline store that does not have any of the plurality of recommended products as a second offline store.

The second offline stores extracted in this way are sorted and searched in order of distance from the location of the first offline store selected by the user. Specifically, scores may be given in the order of the shortest distance to the calculated location of the first offline store, and may be sorted in the order of the highest score.

For example, the second offline stores A, B, and C all have at least one of the plurality of recommended products, and the second offline store A is 0.5 km away from the location of the first offline store, and the second offline store A is 0.5 km away from the location of the first offline store. If the distance of offline store B from the location of the first offline store is 2km, and the distance of the second offline store C from the location of the first offline store is 5km, the server 10 sets the second offline store with the shortest calculated distance. Scores awarded by giving 3 points to store A, 2 points to the second offline store B with the second shortest calculated distance, and 1 point to the second offline store C with the furthest calculated distance. They can be sorted in the order of high second offline stores A, B, and C.

And the three second offline stores sorted based on distance are rearranged according to the order of the number of recommended products in stock among the plurality of recommended products. Specifically, the scores given to each of the second offline stores A, B, and C are weighted according to the number of recommended products in stock to calculate the final score, and the calculated final scores can be rearranged in descending order. .

That is, if 5 recommended products (A, B, C, D, E) are recommended, the second offline store A only has stock for recommended product A among the 5 recommended products, and the second offline store B has If you have stock for recommended products A, B, C, D, and E among the five recommended products, and the second offline store C has stock for recommended products A, B, and C among the five recommended products, , Since the number of recommended products in stock at each of the second offline stores A, B, and C is 1, 5, and 3, weights can be assigned based on this.

For example, the second offline store A is 0.5 km away from the location of the first offline store, the second offline store B is 2 km away from the location of the first offline store, and the second offline store C is 0.5 km away from the location of the first offline store. 1 If the distance from the offline store is 5km, the second offline stores A, B, and C are sorted in that order, and 3 points, 2 points, and 1 point are given to the second offline stores A, B, and C, respectively. The second offline store A has 1 recommended product in stock, the second offline store B has 5 recommended products in stock, and the second offline store C has the number of recommended products in stock. If there are 3, a weight of 3 points is given to the second offline store B with the largest number of recommended products in stock, and a weight of 2 points is given to the second offline store C with the second largest number of recommended products in stock. A weight of 1 point may be assigned to the second offline store A, which has the smallest number of recommended products in stock. Accordingly, since the calculated final scores of the second offline stores A, B, and C are 3 points, 6 points, and 2 points, respectively, they can be rearranged from the order of A, B, and C to the order of B, A, and C. .

At this time, for second offline stores with the same calculated final score, the second offline store with a shorter distance from the location of the first offline store may be determined as priority.

Depending on the embodiment, the server 10 may further consider store business hours when searching for a second offline store. Specifically, the first offline stores may be sorted in order of the most remaining business hours based on the current time.

In the above, it has been explained that when searching for a second offline store, it is sorted based on the distance from the location of the first offline store. However, depending on the embodiment, the second offline store, like the first offline store, is similar to the user's current location. They can be sorted and searched based on distance.

The server 10 may provide the search result (second result) in step S240 to the user terminal (S250).

The user checks the recommended product and the second offline store where the recommended product can be purchased through the user terminal 20, purchases the product that needs to be purchased at the first offline store, and also purchases the product at the second offline store close to the first offline store. You can visit and purchase recommended products. Through this, users can shop offline more efficiently by purchasing the product they need and purchasing a product that matches the product (coordination recommended individually on the service platform) while minimizing movement.

According to the present invention, the first offline store for purchasing products that need to be purchased and the second offline store for purchasing recommended products may be different, but may also be the same. In other words, the user may purchase products that need to be purchased and recommended products, respectively, at the adjacent first offline store and the second offline store, but if the first offline store and the second offline store are the same, the products that need to be purchased and the recommended products are purchased at one offline store. You can also purchase it together.

Meanwhile, the present invention may further include providing, to the user terminal 20, display information of the recommended product in a second offline store selected by the user terminal 20 among the searched second offline stores.

Specifically, the server 10 receives a specific second offline store selected from the user terminal 20 among the searched second offline stores, and provides location information where the recommended product is displayed in the selected specific second offline store to the user terminal 20. By providing this, the user can visit the second offline store and immediately find the recommended product.

At this time, the server 10 may receive the display information in advance from the second offline store terminal 30.

In addition, the present invention may further include receiving, from the user terminal 20, a scheduled visit time for the second offline store selected by the user terminal 20 among the searched second offline stores.

Specifically, a user who has been recommended a second offline store to purchase a recommended product offline can set a scheduled visit time for the second offline store and immediately receive the product to be purchased when visiting the store. For example, a second offline store linked to the personalized recommendation service of the present invention may package products in advance according to a user's visit reservation and then deliver the products through a drive-through method. To this end, users can make payment for recommended products in advance on the service platform.

Depending on the embodiment, the server 10 may provide a stock alarm service to the user. If the recommended product is not in stock at the offline store desired by the user, other than the second offline store searched by the server 10, the user can apply for an alarm service regarding the stock of the recommended product. Afterwards, when the recommended product is stocked at the relevant offline store, a stock alarm is provided in the form of a message to the user terminal 20. At this time, the user enters the expected visit time for the relevant offline store through the user terminal 20. When you visit the store, you will be able to immediately receive recommended products you plan to purchase.

Meanwhile, the present invention provides a step of determining whether there is clothing similar to the recommended product among the clothing held by the user, and if it is determined that there is clothing held similar to the recommended product, a score for coordination between the similar clothing held and the product that needs to be purchased is provided. Additional steps may be included.

The server 10 determines whether there is clothing similar to the recommended product recommended in step S230 among the clothing owned by the user using the learning model built by using the clothing information obtained from each user as big data. can do.

If it is determined that there is no similar clothing, the server 10 may search for a second offline store for the recommended product as described above and provide the recommended product to the user terminal 20 (steps S240 and S250).

On the other hand, if it is determined that there is similar clothing, the server 10 may provide the user terminal 20 with a coordination score that matches the clothing similar to the recommended product to the product that needs to be purchased.

At this time, the coordination score may be a score calculated based on star rating feedback data or satisfaction star rating data input from other users for the coordination in which the recommended product matches the product that needs to be purchased.

If the user's coordination score matching the product that needs to be purchased is high with the clothing that he or she owns that is similar to the recommended product, the user's intention to purchase the product that needs to be purchased may increase, and after purchasing the product that needs to be purchased, the user may hold the product that is similar to the recommended product. You can also wear it by coordinating it with clothing.

In addition, the present invention may further include obtaining image information of the product desired to be purchased from the user terminal and providing coordination recommendations to the user terminal based on the clothing information and image information of the product desired to be purchased. It may be possible.

Specifically, when a user visits an offline store without a plan and finds clothing he likes, the clothing (product desired to be purchased) is photographed through the user terminal 20 and provided to the server 10. ) can provide a coordination recommendation service by extracting clothing that matches the product the user wishes to purchase, based on information on the user's clothing.

Accordingly, the user can determine in advance whether any of the clothing he or she owns matches the product he or she wishes to purchase, and satisfaction with the purchase can be increased by wearing the product he or she wishes to purchase according to the recommended coordination.

Alternatively, the user shows clothing information (image of clothing owned by the user registered in the application) to an employee of an offline store through the user terminal 20, and receives recommendations from the employee for products that match the clothing held in the store. You can also see it.

Alternatively, the user may show the coordination recommended by the service platform to the employee of the offline store through the user terminal 20 and decide to purchase the product matched to the recommended coordination by referring to the employee's opinion about the recommended coordination.

As described above, the present invention provides the value of direct experience through offline shopping and provides various information about the purchaser's purchasing experience in order to resolve the pain point of offline shopping that a lot of time is spent before purchasing a product. By collecting and analyzing, it is possible to present and recommend products that buyers need, as well as information on offline stores where the products can be purchased.

In addition, users can visit an offline store to try on coordinated clothes and purchase them, or find or recommend matching clothes in the store, calculate the coordination score to see if they match the clothes they already own, and determine the degree of fit in advance, saving time on dressing, etc. By minimizing lengthy processes and improving the offline purchasing experience, you can improve consumers' purchase satisfaction and increase repeat visit rates.

2 to 5 describe sequential execution of steps S110 to S224, but this is merely an illustrative explanation of the technical idea of this embodiment, and those skilled in the art will understand As long as it does not deviate from the essential characteristics of this embodiment, various modifications and modifications can be made by executing the order shown in FIGS. 2 to 5 by changing it or executing one or more of steps S110 to S224 in parallel. 2 to 5 are not limited to the time series order.

The method according to an embodiment of the present invention described above may be implemented as a program (or application) and stored in a medium in order to be executed in conjunction with a server, which is hardware.

FIG. 7 is a diagram schematically showing the configuration of a device 500 that performs a method of providing a personalized recommendation service for offline purchases according to the present invention.

Referring to FIG. 7, the processor 510 includes one or more cores (not shown) and a graphics processing unit (not shown) and/or a connection path (e.g., bus, etc.) for transmitting and receiving signals with other components. ) may include.

The processor 510 according to one embodiment executes one or more instructions stored in the memory 520 to perform the method of providing a personalized recommendation service for offline purchases described in relation to FIGS. 1 to 6.

The processor 510 obtains held clothing information related to the held clothing of the user from the user terminal by executing one or more instructions stored in the memory 520, and uses the held clothing based on the held clothing information. Extracting information, extracting a product that needs to be purchased from among the clothes based on the usage information, searching at least one first offline store for the product that needs to be purchased, and sending a first search result to the user terminal. You can perform the steps provided.

The processor 510 executes one or more instructions stored in the memory 520 to analyze characteristic information about the owned clothing based on the held clothing information, and performs a multi-dimensional analysis indicating the degree of similarity between users based on the analysis results. Specifying the user in a space, recommending a product (hereinafter referred to as a recommended product) matching the product to be purchased based on clothing information held in a user group located in a space adjacent to the user in the multidimensional space, Searching for at least one second offline store for a recommended product and providing a second search result to the user terminal may be performed.

Meanwhile, the processor 510 includes random access memory (RAM) (not shown) and read-only memory (ROM) that temporarily and/or permanently store signals (or data) processed within the processor 510. , not shown) may be further included. Additionally, the processor 510 may be implemented in the form of a system on chip (SoC) including at least one of a graphics processing unit, RAM, and ROM.

The memory 520 may store programs (one or more instructions) for processing and controlling the processor 510. Programs stored in the memory 520 may be divided into a plurality of modules according to their functions.

The method of providing a personalized recommendation service for offline purchases according to an embodiment of the present invention described above may be implemented as a program (or application) and stored in a medium to be executed in combination with a computer, which is hardware.

The above-mentioned program is C, C++, JAVA, machine language, etc. that can be read by the processor (CPU) of the computer through the device interface of the computer in order for the computer to read the program and execute the methods implemented in the program. It may include code coded in a computer language. These codes may include functional codes related to functions that define the necessary functions for executing the methods, and include control codes related to execution procedures necessary for the computer's processor to execute the functions according to predetermined procedures. can do. In addition, these codes may further include memory reference-related codes that indicate at which location (address address) in the computer's internal or external memory additional information or media required for the computer's processor to execute the above functions should be referenced. there is. In addition, if the computer's processor needs to communicate with any other remote computer or server in order to execute the above functions, the code uses the computer's communication module to determine how to communicate with any other remote computer or server. It may further include communication-related codes regarding whether communication should be performed and what information or media should be transmitted and received during communication.

The storage medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, examples of the storage medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., but are not limited thereto. That is, the program may be stored in various recording media on various servers that the computer can access or on various recording media on the measurer's computer. Additionally, the medium may be distributed to computer systems connected to a network, and computer-readable code may be stored in a distributed manner.

The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. The software module may be RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Above, embodiments of the present invention have been described with reference to the attached drawings, but those skilled in the art will understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features. You will be able to understand it. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

1: System
10: Server
20: User terminal
30: Offline store terminal
40: communication network
500: device
510: processor
520: memory

Claims (10)

In the method performed by the server,
Obtaining first clothing information related to clothing owned by the user from the user terminal;
Based on the first clothing information, extracting usage information including the number of times the clothing has been worn and the date of purchase;
extracting a product that needs to be purchased from the clothing owned based on the usage information;
extracting at least one first offline store that holds the product to be purchased;
Providing an extraction result of the first offline store to the user terminal;
Analyzing characteristic information about the held clothing based on the first held clothing information;
Specifying the user on a multidimensional space representing the degree of similarity between users based on the analysis results;
Recommending a product (hereinafter referred to as a recommended product) matching the product to be purchased based on second clothing information of a user group located in a space adjacent to the user in the multidimensional space;
extracting at least one second offline store holding the recommended product; and
Comprising: providing the extraction result of the second offline store to the user terminal,
When a plurality of products requiring purchase are extracted, the first offline store extraction step is,
Based on the distance from the user's current location, the extracted first offline stores are sorted in the order of the closest distance to the user, and based on inventory information for each of the plurality of products to be purchased, the sorting is performed. Rearranging the first offline store in the order of the number of products in stock among the plurality of products that need to be purchased,
If multiple recommended products are recommended, the second offline store extraction step is,
Based on the distance from the first offline store, the extracted second offline stores are sorted in the order of the closest distance to the first offline store, and based on inventory information for each of the plurality of recommended products, Rearranging the sorted second offline stores according to the order of the number of recommended products in stock among the plurality of recommended products,
A method of providing personalized recommendation services for offline purchases. delete According to claim 1,
Providing, to the user terminal, display information of the product requiring purchase at a first offline store selected by the user terminal among the extracted first offline stores,
A method of providing personalized recommendation services for offline purchases. According to claim 1,
Further comprising: receiving, from the user terminal, a scheduled visit time for a first offline store selected by the user terminal among the extracted first offline stores,
A method of providing personalized recommendation services for offline purchases. delete According to claim 1,
The characteristic information analysis step is,
extracting at least one clothing factor representing clothing attributes from the first clothing information held; and
Comprising: calculating characteristic values of the user by analyzing characteristic information about the first clothing information held based on the at least one clothing factor,
The user specific step is,
Obtaining coordinate information corresponding to the characteristic value in the multidimensional space; and
Including, specifying the user in the obtained coordinate information,
A method of providing personalized recommendation services for offline purchases. delete According to claim 1,
determining whether there is clothing similar to the recommended product among the clothing owned by the user; and
If it is determined that there is similar clothing, providing a coordination score between the similar clothing and the product that needs to be purchased, further comprising:
A method of providing personalized recommendation services for offline purchases. According to claim 1,
Obtaining image information of a product desired to be purchased from the user terminal; and
Further comprising providing coordination recommendations to the user terminal based on the first clothing information and image information of the product desired to be purchased.
A method of providing personalized recommendation services for offline purchases. A memory that stores one or more instructions; and
It includes a processor that executes the one or more instructions stored in the memory,
The processor, by executing the one or more instructions,
Obtaining first clothing information related to clothing owned by the user from the user terminal;
Based on the first clothing information, extracting usage information including the number of times the clothing has been worn and the date of purchase;
extracting a product that needs to be purchased from the clothing owned based on the usage information;
extracting at least one first offline store that holds the product to be purchased;
Providing an extraction result of the first offline store to the user terminal;
Analyzing characteristic information about the held clothing based on the first held clothing information;
Specifying the user on a multidimensional space representing the degree of similarity between users based on the analysis results;
Recommending a product (hereinafter referred to as a recommended product) matching the product to be purchased based on information on second clothing owned by a user group located in a space adjacent to the user in the multidimensional space;
extracting at least one second offline store holding the recommended product; and
Performing a step of providing the extraction result of the second offline store to the user terminal,
If multiple products requiring purchase are extracted, the processor
Based on the distance from the user's current location, the extracted first offline stores are sorted in the order of the closest distance to the user, and based on inventory information for each of the plurality of products to be purchased, the sorting is performed. Rearranging the first offline store in the order of the number of products in stock among the plurality of products that need to be purchased,
If multiple recommended products are recommended, the processor
Based on the distance from the first offline store, the extracted second offline stores are sorted in the order of the closest distance to the first offline store, and based on inventory information for each of the plurality of recommended products, Rearranging the sorted second offline stores according to the order of the number of recommended products in stock among the plurality of recommended products,
A device that provides personalized recommendation services for offline purchases.