KR101740148B1

KR101740148B1 - Method of recommending items at online shopping malls, based on clients' offline activity data

Info

Publication number: KR101740148B1
Application number: KR1020150153658A
Authority: KR
Inventors: 조혜승; 김경아; 김찬석
Original assignee: 케이티하이텔 주식회사
Priority date: 2015-11-03
Filing date: 2015-11-03
Publication date: 2017-05-25
Also published as: KR20170051931A

Abstract

An online shopping product recommendation method is provided based on offline activity data of a customer who accurately analyzes a customer's interest, inclination, and consumption pattern and recommends a product based on a purchase history of another similar customer. An online shopping product recommendation method based on offline activity data of the customer includes: (a) receiving location information of the customer terminal from the customer terminal; (b) obtaining point of interest (POI) information corresponding to the position information; (c) matching a point of interest (POI) category corresponding to the point of interest (POI) information; (d) establishing a customer segment for customers having the same POI category; (e) analyzing online shopping mall purchase histories of customers constituting the customer segment; And (f) transmitting, from an analysis of the purchase history, recommendation information on a product purchased by another customer in the customer segment to the customer terminal.

Description

A method of recommending online shopping products based on offline activity data of a customer,

The present invention relates to a method for recommending a customized product in online shopping, and more particularly, to a method for recommending an online shopping product based on offline activity data of a customer.

An online shopping mall or an internet shopping mall is a virtual shop that allows users to buy and sell goods using the Internet. Unlike offline shopping malls, online shopping malls have no time and space limitations, so consumers can purchase products at a relatively low price anytime, anywhere. As an operator of an online shopping mall, it is possible to provide merchandise information that meets the taste of the consumer without restriction of the merchandise. The operator of the online shopping mall can directly sell the product to the consumer or relay the transaction by connecting the seller and the consumer. Online shopping has grown rapidly in recent years due to the development of information and communication technology and the convenience of purchasing goods.

On the other hand, the online shopping mall recommends products in various ways in order to promote the purchase activity of the customers. Specifically, two types of analysis methods are adopted for product recommendation. One is analyzing customer's online activity and another is analyzing customer's online activity. An example of the first method is to refer to the purchasing history of the customer and to recommend the product related to the product that was purchased in the past to cause the purchase activity. An example of the second method is recommending a product with a high sales volume, analyzing a comment on a product to recommend a product, or recommending a product with reference to a purchase history of another customer.

Most of the product recommendation methods of online shopping malls suggest products based on the online activities of other people, especially purchase histories. In order for such a product proposal to be successful, there should be a similar interest or shopping pattern between the customer who receives the reference of the purchase history and the customer who receives the product recommendation. However, it is not easy to judge the propensity of two customers or similarity of interest or shopping pattern with online purchase history alone. Because many people use online shopping, but most of them do online purchases only for limited items. Therefore, although online shopping data alone can specify and predict a person's shopping pattern, there is a question about whether the collected and processed data is accurate.

As a result, recommending a product at an online shopping mall in accordance with conventional methods results in a tendency for most customers to be inconsistent with their real interests and thereby ignore the recommendation. In this reality, if an online shopping mall repeatedly recommends a product that does not attract customers' attention, it only causes adverse effects. Online shopper can not like customer's distrust and dislike about repetitive recommendation of online shopping mall.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems and it is an object of the present invention to provide an online shopping product recommendation method based on offline activity data of a customer who accurately analyzes a customer's interest, And the like.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

According to another aspect of the present invention, there is provided a method of recommending an online shopping product based on offline activity data of a customer, comprising the steps of: receiving registered POI information including a place name and location coordinates; A method of recommending an online shopping product using a database storing a grouped interest point category, the service server providing the method comprising: (a) receiving location information of the customer terminal from a customer terminal; (b) obtaining the point of interest (POI) information corresponding to the position information; (c) matching the POI category corresponding to the POI information; (d) setting a customer segment for customers having the same POI category; (e) analyzing online shopping mall purchase histories of the customers constituting the customer segment; And (f) transmitting, from an analysis of the purchase history, recommendation information on a product purchased by another customer in the customer segment to the customer terminal.

The step (b) includes the steps of: collecting coordinates of the customer terminal from the location information; And extracting the point of interest (POI) information corresponding to the coordinates if there is no change in the coordinates for a predetermined period of time.

The point of interest category may include an industry type, and the step (c) may be a step of matching an industry type corresponding to the place name of the POI information.

In the step (d), the customers belonging to the customer segment may have two or more POI categories in common.

In the step (f), when the customers having the predetermined ratio or more with respect to the total number of customers constituting the customer segment purchase the same product, the recommendation information may be generated for the product.

The method may further include executing the shopping application installed in advance by the customer terminal before the step (f), and the recommendation information may be displayed on the customer terminal in association with the shopping application.

The details of other embodiments are included in the detailed description and drawings.

As described above, according to the online shopping product recommendation method based on the offline activity data of the customer according to the present invention, the group of the customers' tendencies or interests is grouped based on the offline activities of the customers, Preference or interest. Customer satisfaction is high for product recommendation information because it refers to purchase history of customers with similar interests.

Specifically, the offline activity data of the customer can be collected using the location information of the customer terminal, and the interest can be accurately identified as the meaningful data based on the offline activity data. For example, if a customer likes a sport, they can collect information about the location of the terminal and analyze it to find out where they go frequently. As a result, we can pinpoint exactly what our customers are interested in, whether they like golf, whether they like swimming or climbing. Likewise, if a customer likes a restaurant, the interest of the customer is changed according to whether the frequent restaurant is a Korean restaurant, a Chinese restaurant, a Western restaurant, or a Japanese restaurant. In this way, if the customer collects data related to where he or she visits offline, and categorizes and analyzes the data, the customer's hobbies and interests can be grasped accurately, which is a fundamental effect of the present invention. However, the advantages of the present invention are not limited to those described above.

People with similar interests have similar shopping patterns. If offline activities are used to group people with similar interests to divide customer segments and recommend products with reference to online purchase histories within each customer segment, the customer can have deep trust in product recommendation information, Purchase activity.

1 is a conceptual diagram briefly illustrating a relationship between a service server and a customer terminal for implementing an online shopping product recommendation method based on offline activity data of a customer according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of a POI-based customer segment management server of FIG. 1. FIG.
3 is a block diagram showing a configuration of the shopping management server of Fig.
4 is a flowchart illustrating an online shopping product recommendation method based on offline activity data of a customer according to an embodiment of the present invention.
FIG. 5 is a flowchart specifically illustrating the POI information extraction step of FIG.
FIG. 6 is an exemplary view illustrating a case where a customer segment is defined by customers sharing one POI category according to an exemplary embodiment of the present invention.
FIG. 7 is a diagram illustrating a case where a customer segment is defined by customers sharing two POI categories according to an exemplary embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

The terms "-subunit" and " -module "in the specification mean units for processing at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software.

The shopping management server according to the present invention generally performs on-line retail distribution related to electronic commerce and can be implemented as an online shopping mall. The online shopping mall can consist of a mall or a specialty mall. A mall is an online shopping mall that sells a variety of product types, and a mall manager can have both a merchandiser function and a profit management function. Because the mall covers various product types, economies of scale can be achieved and a reliable store brand can be formed. The specialty mall is an online shopping mall that is limited to a specific product type, and a professional mall manager can have both a merchandiser function and a profit management function. Since specialized malls handle only products in specific fields, specialized services such as contents, products, and consulting can be provided in the field, thereby securing loyal customers.

The seller in the present invention may include not only a simple seller who sells a product to a consumer but also a producer who actually produces and sells the product or an administrator of a shopping management server that purchases and resells a large amount of goods from a simple seller or a producer have.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a conceptual diagram briefly illustrating a relationship between a service server and a customer terminal for implementing an online shopping product recommendation method based on offline activity data of a customer according to an embodiment of the present invention. 1 is configured to include a POI-based customer segment management server and a shopping management server, FIG. 2 is a block diagram illustrating the configuration of the POI-based customer segment management server of FIG. 1, FIG. 2 is a block diagram showing a configuration of a server. FIG.

The online shopping product recommendation system 10 of the present invention is connected to the customer terminal 100, the seller terminal (not shown) and the wired / wireless network 30 to relay the location analysis and the merchandise sale of the customer terminal 100 And a service server 20 that manages the service.

The customer terminal 100 may include a terminal device having a wireless communication function that a user can move and carry. Examples of such terminals include a mobile communication terminal device, a navigation terminal device, a smart phone, a tablet PC, a wearable smart device, and a PDA capable of communication connection through a mobile communication network. The customer terminal 100 includes a location information providing module 110 and a shopping application 120.

The location information providing module 110 generates current location information of the customer terminal 100 and transmits the location information to the service server 20 through a wired / wireless network 30 such as a mobile communication network system or GPS. The location information may include coordinates on an electronic map and the like.

The shopping application 120 is connected to the service server 20 through a wired / wireless network 30 such as the Internet. The user can access the online shopping mall provided by the service server 20 through the shopping application 120 of the customer terminal 100 to search for and purchase goods. In addition, the shopping application 120 receives the customized product recommendation information from the service server 20 and displays it on the customer terminal 100, thereby assisting the user in purchasing activities.

A customer terminal having a function of transmitting location information to the service server 20 and a customer terminal performing a shopping with the shopping application 120 will be conceptually distinguished from each other. This is because the location information providing module 110 is for providing the location of the mobile terminal to the service server, and the shopping application 120 is for purchasing goods. In the description of the present invention, the former is means for providing basic data for specifying a customer's interest to the service server, and the latter means for receiving the product recommendation analyzed and provided by the service server 20. Therefore, as shown in FIG. 1, such a function may be implemented in one terminal, but the shopping application 120 does not necessarily have to be implemented in the same terminal device as the location information providing module 110.

The shopping application 120 may be implemented in a terminal device having a wired or wireless communication function, which is connected to the service server 20 through the wired / wireless network 30 to transmit and receive information about purchase of goods. For example, a computer, a notebook, a digital TV, a smart phone, a tablet PC, a wearable smart device, a PDA, and an Internet (IoT) appliance can be used.

The location information providing module 110 and the shopping application 120 are implemented in separate terminal devices. A service server 20 that obtains location information of a customer from a first client terminal, such as a smart phone, It is possible to transmit the product recommendation information to the same second customer terminal. The first and second customer terminals may be physically separated as described above. However, in the present specification, it will be collectively referred to as a customer terminal in that it functions to transmit / receive data to / from the service server 20 on behalf of a user. Thus, as shown in FIG. 1, the location information providing module 110 and the shopping application 120 are implemented in a single terminal device.

Although not shown, the merchant terminal is a wired or wireless communication device connected to the shopping management server 300 through the wired / wireless network 30 to transmit and receive information on merchandise sales, and may be, for example, a computer, Smart phones, tablet PCs, wearable smart devices, PDAs, and Internet (IoT) appliances. Merchant registration information, merchandise registration information, order information, and the like can be transmitted and received through the seller terminal. Specifically, the seller terminal transmits product information on a product, a price, and a quantity to be sold in the online shopping mall of the shopping management server 300 to the shopping management server 300. This product information is stored in the product information DB 364 of the shopping management server 300. [

The service server 20 includes a POI (Point of Interest) server 200 for classifying customers having the same interests into customer segments by using the location information of the customer terminal 100, And a shopping management server 300 that provides merchandise recommendation information to customers having interests. The service server 20 may be configured to include a plurality of hardware and / or software devices for database management, data storage and processing, data transmission and reception, authentication and security operations, and the like. In the present embodiment, the service server 20 is implemented as a POI-based POI-based customer segment management server 200 and a shopping management server 300. However, the present invention is not limited to this, 20 may be implemented as a single server. In addition, the POI-based customer segment management server 200 and the shopping management server 300 may each be configured to include one or more hardware and / or software equipment, for example, each of a plurality of servers.

Specifically, the POI-based customer segment management server 200 includes a first control unit 210, a position information receiving unit 220, a POI information extracting unit 230, a POI category matching unit 240, a customer segment setting unit 250 ), And a first storage unit 260. Each component constituting the POI-based customer segment management server 200 may be implemented in one processor or in a separate processor, and may be implemented in a central processing unit (CPU), other chipset, microprocessor, or the like.

The location information receiving unit 220 receives the location information of the current customer terminal 100 from the customer terminal 100.

The POI information extracting unit 230 collects the coordinates of the customer terminal 100 from the position information and extracts POI information corresponding to the coordinates from the POI information DB 262. Here, the POI information DB 262 is a storage medium for storing the point of interest information, and the point of interest information is displayed on the electronic map with the coordinates of the main facility, station, airport, terminal, hotel, restaurant, golf course, Data. For example, the point of interest information may include a place name and coordinates.

The POI information extracting unit 230 may extract the point of interest information for all the coordinates of the customer terminal 100 or may extract the point of interest information only under certain conditions. For example, if the coordinates of the customer terminal 100 vary from moment to moment, it is determined that the customer terminal 100 is moving. However, if the coordinates of the customer terminal 100 do not change for a predetermined period of time, it is determined that the user is performing a specific activity in the corresponding coordinates, so that the POI information extraction unit 230 can extract the point of interest information from the coordinates .

The POI category matching unit 240 extracts a POI category corresponding to the POI information from the POI metadata DB 264 and matches the POI information. Here, the POI metadata DB 264 is a storage medium for storing metadata related to the point of interest information, and the POI metadata may include a place name, a business name, an address, a telephone number, a POI category, and the like. Here, the POI category may include information indicating an attribute of a facility at the place, such as business type. The POI category matching unit 240 of the present embodiment extracts the POI category, i.e., the business type information corresponding to the interest point information when the place name of the POI metadata matches the place name of the POI metadata, and matches the POI category information.

The customer segment setting unit 250 sets a customer segment for customers having at least one POI category. Preferably, a customer segment may be established for customers having more than one POI category. In order to define one customer segment, at least one same POI category is required among customers. The more the same POI category is, the more similar the hobbies, tendencies, interests or shopping patterns of customers in the customer segment tend to be.

The first storage unit 260 includes a POI information DB 262 for storing POI information, a POI metadata DB 264 for storing POI metadata such as a POI category, A customer specific POI DB 266 for storing information, and a customer segment DB 268 for storing customer segment information. In addition, the first storage unit 260 may store programs, contents, and other data such as O / S (Operating System) and various applications necessary for driving the POI-based customer segment management server 200. The first storage unit 260 may be an optical disk such as a hard disk, a CD-ROM, a CD-RW disk, a DVD-ROM, a DVD-RAM, a DVD- Flash memory, various types of RAM, and the like.

The first control unit 210 controls the operation of internal components of the POI-based customer segment management server 200.

Meanwhile, the shopping management server 300 provides information on purchase of goods through an online shopping mall website. The shopping management server 300 analyzes the online shopping mall purchase history of the customers constituting one customer segment by using the customer segment information provided from the POI based customer segment management server 200 and transmits the product recommendation information to the customer terminal 100 send. Also, order information received from the customer terminal 100 is transmitted to a seller terminal (not shown) to mediate / manage the commodity sales. The shopping management server 300 includes a second control unit 310, a transmission / reception unit 320, a purchase pattern analysis unit 330, a product recommendation unit 340, and a second storage unit 360. Each component configuring the shopping management server 300 may be implemented in a single processor or in a separate processor, or in a central processing unit, other chipset, microprocessor, or the like.

The transceiver 320 is connected to the customer terminal 100 and the seller terminal (not shown) through the wired / wireless network 30 to transmit and receive information.

The purchase pattern analyzing unit 330 analyzes online shopping mall purchase histories of customers belonging to the same customer segment based on the customer segment information provided from the POI based customer segment management server 200. [ For example, the purchase pattern analyzer 330 can set the order of the popular items in the order of the number of orders, among the products that the customers have purchased in the past. Since the customers in the same customer segment have similar tendencies, interests or shopping patterns, it is possible to increase the purchase rate of the recommended product when recommending the popular product.

The product recommendation unit 340 transmits the product recommendation information to the customer terminal 100 for the popular product set by the purchase pattern analysis unit 330. Preferably, the product recommendation unit 340 may generate recommendation information for the product if the customers who purchase the same product at a predetermined ratio of the total number of customers constituting the customer segment purchase the same product.

The second storage unit 360 includes a customer information DB 362 that stores customer registration information and seller registration information, a product information DB 364 that stores product information, a purchase history DB 364 that stores customer purchase history, (366), and a recommended product DB 368 for storing recommended product information for each customer segment. In addition, the second storage unit 360 may store programs, contents, and other data such as O / S and various applications required for driving the shopping management server 300. The second storage unit 360 may be an optical disk such as a hard disk, a CD-ROM, a CD-RW disk, a DVD-ROM, a DVD-RAM, a DVD- Flash memory, various types of RAM, and the like.

The second control unit 310 controls the operations of the internal components of the shopping management server 300.

4 and 5, an online shopping product recommendation method based on offline activity data of a customer according to an embodiment of the present invention will be described. FIG. 4 is a flowchart illustrating an online shopping product recommendation method based on offline activity data of a customer according to an embodiment of the present invention, and FIG. 5 is a flowchart illustrating the POI information extraction step of FIG.

First, the customer of the customer terminal 100 executes a web browser and can input the address of the online shopping mall website operated by the shopping management server 300. The customer terminal (100) accesses the shopping management server (300) based on the inputted address and receives the online shopping screen data. Accordingly, the customer terminal 100 provides a shopping screen to the customer. The customer enters his / her information using the customer terminal 100 and joins the shopping management server 300 as a customer member. Likewise, the seller enters his / her information using a seller terminal (not shown) and enters the seller's membership into the shopping management server. The seller can register the product information including the product, price, quantity, origin, manufacturing source, expiration date, etc. of the product by using the seller terminal in the shopping management server. The POI-based customer segment management server 200 can access the customer registration information stored in the customer information DB 362 and the POI-based customer segment management server 200 can recognize the information of the customer terminal 100 .

The location information providing module 110 of the customer terminal 100 generates the current location information of the customer terminal 100 and transmits the location information to the POI based customer segment management server 200 through the wired and wireless network 30 ). Here, the position information may include coordinates on an electronic map and the like.

The POI information extraction unit 230 of the POI based customer segment management server 200 collects the coordinates of the customer terminal 100 from the position information and extracts the point of interest information corresponding to the coordinates from the POI information DB 262 (S12). 5, when the location information receiving unit 220 receives the location information from the customer terminal 100 (S121), the POI information extracting unit 230 extracts the location information from the location information of the customer terminal 100, (S122). Then, the POI information extracting unit 230 determines whether the coordinates have changed for a predetermined time (S123). If there is a coordinate change for a predetermined time, the POI information extracting unit 230 determines that the customer is moving and continuously collects the coordinates without extracting the point of interest information from the coordinates. If there is no coordinate change for a predetermined time, the POI information extracting unit 230 determines that the customer is performing a specific activity in the coordinates and extracts the point of interest information corresponding to the coordinates from the POI information DB 262 S124).

Then, the POI category matching unit 240 extracts a POI category corresponding to the POI metadata from the POI metadata DB 264, and matches the POI category with the POI information (S20). Here, the point of interest category may indicate the property information on the facility of the corresponding indicator, for example, an industry type. According to an embodiment of the present invention, the location information is converted into various points of interest information by a long-term offline activity of the customer, the point of interest information is classified according to the location property, and the POI category is matched for each classification.

The customer segment setting unit 250 sets and sets a customer segment by classifying and grouping customers having the same POI category (S30).

The POI-based customer segment management server 200 transmits the customer segment information to the shopping management server 300 (S32).

The purchase pattern analysis unit 330 of the shopping management server 300 analyzes online shopping mall purchase histories of customers belonging to the same customer segment based on the customer segment information at step S40. For example, popular product information that is repeatedly purchased by customers in the same customer segment is extracted to analyze the purchase patterns of the customers.

Subsequently, the product recommendation unit 340 generates product recommendation information for the popular product (S42), and transmits the product recommendation information to the customer terminal 100 (S52).

Preferably, the customer terminal 100 may call the shopping application 120 before receiving the product recommendation information (S50). The shopping application 120 can display the product recommendation information on the terminal display device.

The customer can access the online shopping mall of the shopping management server 300 through the shopping application 120 and use the product recommendation information when searching for and purchasing a product. Since the product recommendation information provided in this way is generated by referring to the purchase history of another customer having a similar interest or life pattern, it can be used as very useful information. The customer can transmit the order information of the goods to the shopping management server 300 through the shopping application 120 (S54). The seller then delivers the goods to the customer according to the order information.

Hereinafter, the correlation between the POI information, the POI category, and the customer segments of the present invention will be described with reference to FIGS. 6 and 7. FIG. At least one POI information is generated from the location information of the customer terminal 100. The at least one POI information is grouped according to the POI category. Thus, at least one POI category is defined for a single customer. Since the customer segment is set for at least one POI category, the customer can belong to at least one customer segment.

FIG. 6 is an exemplary view illustrating a case where a customer segment is defined by customers sharing one POI category according to an exemplary embodiment of the present invention. The POI information of the customer Sean is classified into four POI categories. In this example, the POI category information of the customer Sean is A (shopping), B (mountain climbing), C (golf), and D (restaurant). Daniel's POI information is classified into four POI categories. In this example, Daniel's POI category information is C (golf), D (restaurant), E (swimming), and F (theater). The POI information of the customer Sam is classified into three POI categories. In this example, the POI category information of the customer Sam is A (shopping), D (restaurant), and G (library).

The customers (Sean, Sam) who constitute the first customer segment have a common interest in the A (shopping) POI category. Customers who make up the second customer segment (Sean, Daniel) have a common interest in the C (golf) POI category. Customers who make up the third customer segment (Sean, Daniel, Sam) have a common interest in the D (restaurant) POI category.

As described above, since customers who have a common interest in the customer segment refer to each other's online shopping mall purchase history, the customer satisfaction with the product recommendation information can be increased.

FIG. 7 is a diagram illustrating a case where a customer segment is defined by customers sharing two POI categories according to an exemplary embodiment of the present invention. The customers (Sean, Sam) who make up the first customer segment have a common interest in the A (shopping) and D (restaurant) POI categories. Customers who make up the second customer segment (Sean, Daniel) have a common interest in the C (golf) and D (restaurant) POI categories. Compared with the example of FIG. 6, in the case of the example of FIG. 7, product recommendation information is more effective because it shares more interests among customers in one customer segment.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

10: online shopping product recommendation system 20: service server
30: wired / wireless network 100: customer terminal
110: location information providing module 120: shopping application
200: POI-based Customer Segment Management Server
210: first control unit 220: position information receiving unit
230: POI information extracting unit 240: POI category matching unit
250: customer segment setting unit 260: first storage unit
262: POI information DB 264: POI metadata DB
266: Customer-specific POI DB 268: Customer Segment DB
300: shopping management server 310: second control unit
320: Transmitting / receiving unit 330: Purchase pattern analyzing unit
340: product recommendation unit 360: second storage unit
362: customer information DB 364: product information DB
366: Purchase history DB 368: Recommended product DB

Claims

A method of recommending an online shopping product using registered point of interest (POI) information including a name of a place and a location, and a database storing interest point category information in which the point of interest information is grouped,
A service server providing the above method comprises:
(a) receiving location information of the customer terminal from a customer terminal;
(b) obtaining point of interest (POI) information corresponding to the location information;
(c) matching a point of interest (POI) category corresponding to the point of interest (POI) information;
(d) establishing the same customer segment for customers having the same POI category;
(e) analyzing online shopping mall purchase histories of customers belonging to the same customer segment; And
(f) transmitting, from the analysis of the purchase history, recommendation information on a product purchased by another customer in the same customer segment to the customer terminal,
Wherein the customer segment is a group consisting of customers whose interest points (POI) obtained from the location information of each customer's terminal belong to the same point of interest (POI) category, based on the offline activity data of the customer .

The method of claim 1, wherein the step (b)
Collecting coordinates of the customer terminal from the location information; And
And extracting the point of interest (POI) information corresponding to the coordinates when the coordinates do not change for a predetermined period of time.

3. The method of claim 2,
The interest point category includes an industry category,
Wherein the step (c) is a step of matching an industry corresponding to the place name of the POI information.

The method of claim 1, wherein, in step (d)
Wherein the customers belonging to the customer segment are based on offline activity data of a customer having two or more POI categories in common.

The method according to claim 1, wherein in the step (f)
Based on offline activity data of a customer for which the recommendation information is generated for the product when a customer of a predetermined ratio or more with respect to the total number of customers constituting the customer segment has purchased the same product.

The method according to claim 1,
Further comprising the step of executing a shopping application previously installed by the customer terminal before the step (f)
And the recommendation information is based on offline activity data of a customer displayed on the customer terminal in association with the shopping application.