WO2022005140A1 - Procédé, dispositif et système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré - Google Patents

Procédé, dispositif et système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré Download PDF

Info

Publication number
WO2022005140A1
WO2022005140A1 PCT/KR2021/008120 KR2021008120W WO2022005140A1 WO 2022005140 A1 WO2022005140 A1 WO 2022005140A1 KR 2021008120 W KR2021008120 W KR 2021008120W WO 2022005140 A1 WO2022005140 A1 WO 2022005140A1
Authority
WO
WIPO (PCT)
Prior art keywords
customer
personal information
product
shopping mall
information element
Prior art date
Application number
PCT/KR2021/008120
Other languages
English (en)
Korean (ko)
Inventor
노동우
최원규
문현규
김명구
정인숙
황소연
Original Assignee
카페24 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from KR1020200079824A external-priority patent/KR102538398B1/ko
Priority claimed from KR1020200079823A external-priority patent/KR102518389B1/ko
Priority claimed from KR1020200079822A external-priority patent/KR20220001616A/ko
Application filed by 카페24 주식회사 filed Critical 카페24 주식회사
Publication of WO2022005140A1 publication Critical patent/WO2022005140A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions

Definitions

  • the present invention relates to a method, apparatus and system for constructing customer big data based on the generation of an integrated customer identifier, and more particularly, to integrated management of customer information scattered in a plurality of shopping malls based on e-commerce. It relates to a method, apparatus and system for building customer big data based on the generation of an integrated customer identifier that can generate an integrated customer identifier and build big data based on it.
  • marketing is an essential success factor in e-commerce based on the Internet based on the Internet.
  • banner advertisements or image advertisements are exposed on web pages such as portal sites or e-commerce sites that provide e-mail services, search services, and e-commerce services, and the user clicks the advertisements. If you do, you will be connected to the linked advertiser's server.
  • advertisement content is provided to an unspecified number of Internet users without distinction of residence area, age, and taste, etc. There are also minor problems.
  • Korean Patent Laid-Open No. 2013-0157475 discloses a reference information collection unit that provides an advertisement environment setting interface to a user and collects advertisement environment setting information from the user and the collected advertisements.
  • a technology such as a target advertisement providing apparatus including an advertisement management unit that detects candidate advertisements based on environment setting information and selects and organizes exposure advertisements for each advertisement slot from among the detected candidate advertisements.
  • the conventional target advertisement providing technology uses a method in which a link for collecting customer information is inserted into an internal frame (I-Frame) in each web page to be connected to the user terminal and the user information collection server, but this is It is not efficient in terms of load, and since each shopping mall operator uses a different information collection method, the collected information is limited and the amount of data is small, so the accuracy is lowered.
  • I-Frame internal frame
  • the present invention is to create an integrated customer identifier that enables the integrated management of customer activity information scattered in a plurality of e-commerce-based shopping malls, and builds big data based on this for efficient marketing. It aims to provide a method, device and system for building customer big data based on the generation of integrated customer identifier that can be utilized.
  • the present invention provides a customer big data construction method in one aspect (Aspect).
  • the customer big data construction method is performed by a server interworking with a plurality of e-commerce-based shopping malls, and from the plurality of shopping malls, a plurality of individual shopping mall customer identifiers corresponding to the plurality of shopping malls and the plurality of individual shopping mall customer identifiers
  • Receiving a plurality of personal information element groups corresponding to, -Personal information elements included in the plurality of personal information element groups are de-identified personal information; specifying a first customer based on the plurality of groups of personal information elements; generating a unified customer identifier corresponding to the first customer; storing the integrated customer identifier corresponding to the first customer in association with the plurality of individual shopping mall customer identifiers and the plurality of personal information element groups in a database; and collecting log records of the first customer from the plurality of shopping malls based on the integrated customer identifier.
  • the step of specifying the first customer may include: calculating a degree of personal information similarity between the plurality of personal information element groups; and specifying the first customer based on the calculated similarity of personal information.
  • the calculating of the personal information similarity may include calculating the number of personal information elements included in the personal information element group and the number of personal information elements matching between the personal information element groups.
  • the step of specifying the first customer based on the personal information similarity may include specifying the same customer if the calculated personal information similarity is greater than or equal to a predetermined value.
  • the step of specifying the customer may include inputting data of a plurality of personal information element groups into an artificial intelligence engine learned through deep learning and classifying by the same customer.
  • the de-identified personal information includes customer email, customer name, and customer phone number. It may be associated with at least two or more of a customer nickname, customer address, customer date of birth, customer gender, and customer age.
  • the method for constructing customer big data includes receiving, from a first shopping mall, a first individual shopping mall customer identifier capable of identifying a second customer in the first shopping mall and a first personal information element group corresponding to the first individual shopping mall customer identifier to do; determining whether an integrated customer identifier corresponding to the second customer exists in the database based on the first individual shopping mall customer identifier; generating a unified customer identifier corresponding to the second customer based on the first personal information element group when the unified customer identifier corresponding to the second customer does not exist in the database;
  • the method may further include the step of associating a unified customer identifier corresponding to the second customer with the first individual shopping mall customer identifier and the first personal information element group and storing in the database.
  • the customer big data construction method includes comparing the first personal information element group with the personal information element group of the second customer stored in the database when there is an integrated customer identifier corresponding to the second customer in the database Based on that, determining whether there is a changed content; and if there is a changed content as a result of the comparison, updating the content of the second customer's personal information element group based on the changed content may further include.
  • the customer big data construction method may include, based on the collected log records of the first customer, a product recommendation corresponding to the first customer based on at least one of a product recommendation based on a graph database and a basket recommendation It may further include the step of performing.
  • the customer big data construction apparatus may include: a customer big data construction apparatus in an e-commerce system, comprising: a communication unit communicating with at least one of a wired network and a wireless network; receiving a plurality of individual shopping mall customer identifiers corresponding to the plurality of shopping malls and a plurality of personal information element groups corresponding to the plurality of individual shopping mall customer identifiers from a plurality of shopping malls based on the communication unit;
  • the personal information elements included in the de-identified personal information specify a first customer based on the plurality of personal information element groups, generate a unified customer identifier corresponding to the first customer, and
  • the integrated customer identifier corresponding to one customer is stored in a database by associating the plurality of individual shopping mall customer identifiers and the plurality of personal information element groups, and based on the integrated customer identifier, the first customer from the pluralit
  • the processor may calculate the personal information similarity between the plurality of personal information element groups, and specify the first customer based on the calculated personal information similarity.
  • the processor may calculate the number of personal information elements included in the personal information element group and the number of personal information elements matching between the personal information element groups.
  • the processor may specify the same customer as the calculated similarity of personal information is greater than or equal to a predetermined value.
  • the processor may input data of a plurality of personal information element groups into an artificial intelligence engine learned through deep learning and classify them by the same customer.
  • the de-identified personal information includes customer email, customer name, and customer phone number. It may be associated with at least two or more of a customer nickname, customer address, customer date of birth, customer gender, and customer age.
  • the processor is configured to receive, from a first shopping mall, a first individual shopping mall customer identifier capable of identifying a second customer in the first shopping mall and a first group of personal information elements corresponding to the first individual shopping mall customer identifier; 1 Based on the individual shopping mall customer identifier, it is determined whether an integrated customer identifier corresponding to the second customer exists in the database, and if the integrated customer identifier corresponding to the second customer does not exist in the database, the first Generating an integrated customer identifier corresponding to the second customer based on one personal information element group, and combining the integrated customer identifier corresponding to the second customer with the first individual shopping mall customer identifier and the first personal information element group It can be associated and stored in the database.
  • the processor is configured to: if there is a unified customer identifier corresponding to the second customer in the database, based on comparing the group of personal information elements of the second customer and the group of personal information elements of the first individual stored in the database, , it is possible to determine whether the contents have been changed. In this case, if there is a changed content as a result of the comparison, the content of the second customer's personal information element group may be updated based on the changed content.
  • the processor may perform a product recommendation corresponding to the first customer based on at least one of a product recommendation based on a graph database and a basket recommendation based on the collected log records of the first customer. have.
  • the present invention provides a customer big data construction system in another aspect.
  • the customer big data construction system a plurality of shopping malls; and receiving a plurality of individual shopping mall customer identifiers corresponding to the plurality of shopping malls and a plurality of personal information element groups corresponding to the plurality of individual shopping mall customer identifiers from the plurality of shopping malls, included in the plurality of personal information element groups personal information elements are de-identified personal information - specify a first customer based on the plurality of groups of personal information elements, generate a unified customer identifier corresponding to the first customer, and assign to the first customer
  • the corresponding integrated customer identifier is associated with the plurality of individual shopping mall customer identifiers and the plurality of personal information element groups and stored in a database, and log records of the first customer from the plurality of shopping malls based on the integrated customer identifier It may include a server that collects
  • the server may calculate the degree of personal information similarity between the plurality of personal information element groups, and specify the first customer based on the calculated degree of personal information similarity.
  • the server may calculate the number of personal information elements included in the personal information element group and the number of matching personal information elements between the personal information element groups.
  • the present invention is performed by a server that interworks with a plurality of e-commerce-based shopping malls in a computer, and a server corresponding to the plurality of shopping malls from the plurality of shopping malls
  • an integrated customer identifier is generated in order to integratedly manage customer activity information scattered in a plurality of e-commerce-based shopping malls, and based on this, a big information related to customer activity in a plurality of shopping mall groups is generated.
  • Data can be collected easily. Therefore, it is possible to utilize the analyzed data for efficient marketing through reliable analysis through big data.
  • FIG. 1 is a block diagram showing the configuration of a system for building customer big data based on generation of integrated customer identifiers and product recommendation based thereon according to a preferred embodiment of the present invention.
  • FIG. 2 is a flowchart illustrating a customer big data construction process according to a preferred embodiment of the present invention.
  • FIG. 3 is a flowchart for explaining an embodiment of a process for specifying a first customer shown in FIG. 2 .
  • FIG. 4 is an exemplary diagram illustrating a table stored and managed in the database 140 .
  • FIG. 5 is a conceptual diagram for explaining a product recommendation method based on similarity.
  • FIG. 6 is a graph based on product, order, and customer nodes used in a product recommendation method using a graph database according to an embodiment of the present invention.
  • FIG. 7 is a conceptual diagram for explaining a direct correlation and an indirect correlation between products.
  • FIG. 8 is a flowchart illustrating a product recommendation method using a graph database according to an embodiment of the present invention.
  • FIG. 9 is a conceptual diagram illustrating a state in which a plurality of product-centered radial graphs are entangled.
  • FIG. 10 is a diagram for explaining the appropriate use of the same/similar category and combination category in a product recommendation method using a graph database according to an embodiment of the present invention.
  • FIG. 11 is a conceptual diagram illustrating an AI learning method related to a recommendation algorithm used in a product recommendation method using a graph database according to an embodiment of the present invention.
  • FIG. 12 is a conceptual diagram for explaining a method of indexing a connection path of a product recommendation method using a graph database according to an embodiment of the present invention.
  • 13 to 16 are diagrams illustrating various embodiments of finding similar products and customers with similar tendencies as outputs by designating product and customer nodes as inputs.
  • 17 is a diagram illustrating a result of implementing a product recommendation method using a graph database in a web page according to an embodiment of the present invention.
  • FIG. 18 is a flowchart illustrating a product recommendation method using a graph database according to another embodiment of the present invention.
  • 19 and 20 are diagrams illustrating exemplary aspects of product route prediction in a product recommendation method using a graph database according to another embodiment of the present invention.
  • 21 and 22 are diagrams illustrating exemplary aspects of product route prediction in a situation in which the next product is inquired after the product inquiry in the embodiment of FIG. 19 .
  • 23 is a conceptual diagram for explaining a situation in which promotion of products whose sales are low according to product sales activity analysis is required.
  • 24 and 25 are conceptual views for explaining a matching method for providing a special promotion by matching a specific number of customer groups and a specific number of product groups when there are 5 customers and 6 products.
  • 26 and 27 are flowcharts illustrating a product basket recommendation method according to another embodiment of the present invention.
  • FIG. 28 is a flowchart illustrating a product-based basket recommendation method according to another embodiment of the present invention. This relates to the method (1) of FIG. 24 .
  • 29 to 31 are tables for describing in detail a method of generating a matrix for each step of the basket recommendation method of FIG. 28 .
  • FIG. 32 is a flowchart illustrating a customer-based basket recommendation method according to another embodiment of the present invention.
  • 33 to 35 are tables for describing in detail a method of generating a matrix for each step of the basket recommendation method of FIG. 32 .
  • 36 is an apparatus for constructing customer big data based on generation of an integrated customer identifier according to an embodiment of the present invention, and performing a product recommendation method and/or a product basket recommendation method based on a graph database based on the constructed big data is a schematic block diagram of
  • first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.
  • FIG. 1 is a block diagram showing the configuration of a system for building customer big data based on generation of integrated customer identifiers and product recommendation based thereon according to a preferred embodiment of the present invention.
  • the system may include a plurality of shopping malls 110 - 1 to 110 -N, a network 120 , a server 130 , a database 140 , and a user terminal 150 .
  • the user terminal 150 may be a customer terminal using a shopping mall.
  • the server 130 may interwork with a plurality of e-commerce-based shopping malls 110-1 to 110-N through the network 120 .
  • Each of the shopping malls 110-1 to 110-N may be, for example, a web server providing an e-commerce site. That is, the shopping malls 110-1 to 110-N refer to online shopping malls that sell products or services through the Internet.
  • the shopping malls 110-1 to 110-N may have a specific URL and sell products to visitors to the corresponding address on the Internet.
  • the shopping mall operator operates the shopping mall in such a way that at least a portion of the amount of the sold product is acquired through his/her own account associated with the shopping mall.
  • the shopping malls 110-1 to 110-N may be not only a clothing shopping mall, an open market, etc., but also a game site, social media site, community site, broadcasting site, multimedia providing site, etc. that sell goods or services for a fee. have.
  • the shopping malls 110-1 to 110 to N are, as described above, a web page related to a specific address (eg, URL) on the Internet, a server or a shopping mall that handles information related to the shopping mall It can be implemented as a terminal of the operator of The terminal may include a wired/wireless networking computing device such as a smart phone or a PC.
  • the plurality of shopping malls 110-1 to 110-N may be operated by different types of companies.
  • the first shopping mall 110-1 may be a web server for providing an Internet shopping mall operated by e-commerce company A
  • the second shopping mall 110-2 is completely unrelated to e-commerce company A
  • B may be a service web server for providing an Internet shopping mall operated by an e-commerce company.
  • the plurality of shopping malls 110 - 1 to 110 -N may interwork with the server 130 through the network 120 .
  • the server 130 is a device that obtains information from a plurality of shopping malls 110-1 to 110-N and recommends a product in response to a product recommendation query.
  • the server 130 may be a server (not shown) that distributes and manages a platform supporting the opening and operation of a shopping mall, or a server interworking with the server.
  • the server 130 according to a preferred embodiment of the present invention does not handle only information related to one shopping mall 110-1 based on generating and managing the integrated customer identifier, but a plurality of shopping malls 110-1 to 110-N) and related information, and can analyze and learn various information based on big data.
  • the server 130 may be implemented as one or a plurality of computing devices.
  • the computing device referred to herein may be implemented as a server-class computer terminal.
  • the computing device may include an input device, a display device, a networking device, a hard disk, a memory for storing a program, and a processor for executing a program stored in a typical computer terminal. However, it is not necessarily implemented as a server-grade computer terminal.
  • the server 130 generates an integrated customer identifier that can identify the same customer in the plurality of shopping malls 110-1 to 110-N, and uses the integrated customer identifier from the plurality of shopping malls 110-1 to 110-N according to the integrated customer identifier. It is possible to collect log records of identified customers, and perform marketing through product recommendations tailored to customers based on the collected log records.
  • the collection of a user's Internet use history is performed for each individual e-commerce company.
  • an individual e-commerce company records the cookie or session of the user who accessed the shopping mall operated by the e-commerce company, and then records the user's previous usage history when the user accesses the shopping mall of the e-commerce company again.
  • a product advertisement is displayed on a web page.
  • the target advertisement performed by each e-commerce company as described above, the amount of collected data is small and the reliability is low because the data on the customer's usage history is based on the e-commerce usage history for one shopping mall.
  • the server 130 includes a plurality of individual shopping mall customer identifiers and Receive a plurality of personal information element groups corresponding to a plurality of individual shopping mall customer identifiers, specify a customer based on the plurality of personal information element groups to generate an integrated customer identifier corresponding to the specified customer, and respond to the customer
  • a plurality of shopping malls 110-1 to 110-N Big data can be built by collecting log records of specified customers from
  • the integrated customer identifier may be a customer identifier for identifying the same customer as one identifier in the plurality of shopping malls 110-1 to 110-N.
  • the network 120 includes a plurality of shopping malls 110-1 to 110-N and a server 130, a server 130 and a user terminal 150, and a plurality of shopping malls 110-1 to 110-N and a user terminal ( 150) may mean a communication network that connects them.
  • the network 120 may include a wired network or a wireless network (the Internet).
  • FIG. 2 is a flowchart illustrating a customer big data construction process according to a preferred embodiment of the present invention.
  • the server 130 may access a plurality of shopping malls 110-1 to 110-N.
  • the server 130 provides a plurality of individual shopping mall customer identifiers corresponding to the plurality of shopping malls 110-1 to 110-N from the plurality of shopping malls 110-1 to 110-N and a plurality of individual shopping mall customer identifiers corresponding to the plurality of individual shopping mall customer identifiers.
  • a group of personal information elements may be received (S1).
  • the individual shopping mall customer identifier may be a customer identifier used to identify a customer in the individual shopping mall.
  • the personal information element group may mean personal information elements corresponding to the individual shopping mall customer identifier. That is, it may mean a set of personal information elements registered and managed in an individual shopping mall in response to an individual customer identifier.
  • the first shopping mall 110 - 1 identifies the first customer with respect to the first customer based on the individual shopping mall customer identifier corresponding to the first customer, and in response to the individual shopping mall customer identifier, the first customer You can manage elements of your personal information.
  • the personal information elements included in the personal information element group may be non-identified personal information elements.
  • the de-identified personal information element may refer to a personal information element other than information used to identify a customer in the shopping mall, such as an identifier (ID) and a resident registration number.
  • the de-identified personal information elements include customer email, customer name, customer phone number. This may be a customer nickname, customer address, customer date of birth, customer gender, customer age, and the like.
  • the server 130 may specify the first customer based on the plurality of personal information element groups (S2). This can be said to specify a customer by comparing the non-identified personal information element group corresponding to each shopping mall 110-1 to 110-N with each other and viewing a customer having a high similarity personal information element group as the same customer. .
  • FIG. 3 is a flowchart for explaining an embodiment of a process for specifying a first customer shown in FIG. 2 .
  • the server 130 may calculate the degree of personal information similarity between the plurality of personal information element groups ( S11 ). For example, the server 130 may calculate the number of personal information elements included in the personal information element group and the number of personal information elements that match between the personal information element groups.
  • the server 1230 may specify the first customer based on the calculated similarity of personal information. For example, the server 140 determines whether the calculated personal information similarity is equal to or greater than a predetermined value (S12), and when the calculated personal information similarity is greater than or equal to a predetermined value, the server 140 specifies the same customer (S13).
  • S12 a predetermined value
  • S13 the same customer
  • the number of personal information elements included in one personal information element group is 10, and the personal information similarity value determined to be the same customer, that is, the reference value is 0.7, there are 7 personal information elements that match each other
  • a customer corresponding to the above group of personal information elements may be determined as the same customer and specified as one customer.
  • the server 130 may input the received data of the plurality of personal information element groups into an artificial intelligence engine learned through deep learning and classify them by the same customer.
  • the artificial intelligence engine classifies personal information element groups with a high degree of personal information similarity to each other and responds to the same customer, so that it is learned to classify by the same customer. have.
  • the server 130 may generate an integrated customer identifier corresponding to the first customer specified above (S3).
  • the integrated customer identifier may be a customer identifier for identifying the same customer as one identifier in the plurality of shopping malls 110-1 to 110-N, as described above.
  • the server 130 may associate the customer identifiers corresponding to the generated first customer with the plurality of individual shopping mall customer identifiers and the plurality of personal information element groups and store them in the database 140 (S4).
  • FIG. 4 is an exemplary diagram illustrating a table stored and managed in the database 140 .
  • the first customer is a customer of shopping mall A, shopping mall B, shopping mall C, and shopping mall D
  • each shopping mall uses different individual shopping mall customer identifiers. That is, “AAAA” in shopping mall A, “BBBB” in shopping mall B, “CCCC” in shopping mall C, and “DDDD” in shopping mall D are used as “Mall-ID” as individual shopping mall customer identifiers.
  • each shopping mall corresponds to a group of personal information elements including de-identified personal information elements, for example, email, phone number, name, nickname, age, gender, etc. of the first customer.
  • the server 130 determines that "AAAA” of shopping mall A, "BBBB” of shopping mall B, "CCCC” of shopping mall C, and “DDDD” of shopping mall D are the same as the first customer. 4, by generating “CUAA01” as an integrated customer identifier “EC-ID” capable of identifying the first customer in the four shopping malls, and associating it with a plurality of individual shopping mall customer identifiers and personal information element groups, as shown in FIG.
  • the same table as in the above example may be stored and managed in the database 140 .
  • the server 13 sends the first shopping mall 110-1 to the first shopping mall ( In 110-1), an individual shopping mall customer identifier capable of identifying the second customer and a personal information element group corresponding to the individual shopping mall customer identifier of the second customer may be received.
  • the server 130 may determine whether an integrated customer identifier corresponding to the second customer exists in the database 140 based on the individual shopping mall customer identifier of the second customer. Here, since the second customer is a newly introduced customer, the integrated customer identifier of the second customer does not exist in the database 140 .
  • the server 130 is configured to respond to the second customer based on the group of personal information elements corresponding to the second customer.
  • the integrated customer identifier may be generated, and the integrated customer identifier corresponding to the second customer may be associated with the individual shopping mall customer identifier of the second customer and the personal information element group corresponding to the individual shopping mall customer identifier and stored in the database 140 .
  • the server 130 performs the Based on comparing the second customer's personal information element group stored in the database 140 with the received personal information element group, it is determined whether there is any changed content, and if there is changed content as a result of the comparison, based on the changed content may update the content of the second customer's personal information element group.
  • the server 130 records the log records of the first customer from the plurality of shopping malls 110-1 to 110-N based on the integrated customer identifier stored and managed in the database 140 as described above. can be collected (S6). That is, big data related to customers' order histories, such as purchase history, connection history, and return history, performed in the plurality of shopping malls 110-1 to 110-N for each customer may be collected.
  • the server 130 collects the collected customer order history data to build big data, and then a customized product corresponding to the customer based on at least one of a product recommendation by a first method and a product recommendation by a second method Recommendations can be performed.
  • the product recommendation according to the first method may be, for example, a product recommendation based on a similarity using a graph database.
  • the product recommendation according to the second method may be a product recommendation between a group customer and a group product, for example, a basket recommendation.
  • the first method is a product recommendation based on similarity using a graph database.
  • the degree of similarity is different from the degree of similarity of personal information between a plurality of personal information element groups calculated when generating the aforementioned integrated customer identifier.
  • the degree of similarity using the graph database will be described in detail below.
  • the server 130 receives information from a shopping mall operation server (not shown) that generally manages information on shopping malls 110-1 to 110-N and/or shopping malls 110-1 to 110-N. Obtain order history data and/or product inquiry data. In the data, a customer is identified by a unified customer identifier.
  • the server 130 converts the obtained data into a graph form having a plurality of node layers.
  • the data converted to the graph form is stored in the database 140 external to the server 130 .
  • the database 140 does not necessarily exist outside the server 130 . It may exist inside.
  • the server 130 responds to a product recommendation request from the user terminal 150 (which may be referred to as a “recommended product search request”, “query”, etc.) from a specific product by using a graph composed of three node hierarchies.
  • a product recommendation request from the user terminal 150 (which may be referred to as a “recommended product search request”, “query”, etc.) from a specific product by using a graph composed of three node hierarchies.
  • Process queries for search At this time, on the graph data in the converted form, a connection path between a product and a product, a customer and a customer, or a product and a customer is extracted, and the similarity between each other can be determined by considering the distance and/or the number of the extracted connection paths. .
  • queries on similar products are processed by analyzing the inquiry history for a certain step before or after product search on the graph. The part related to the technology for processing queries for similar products will be described in more detail below.
  • the server 130 uses an indicator (which may include a similarity using the graph database above) indicating the purchaseability between the individual customer and the individual product based on the existence of the product. Baskets can be optimized. That is, by using the product basket recommendation method according to the embodiment of the present invention, it is possible to appropriately promote a plurality of product groups optimized for at least one customer (in a plurality of cases, a customer group).
  • the user terminal 150 may be a terminal of a purchaser of the shopping mall 110 - 1 .
  • the purchaser may transmit a product recommendation request to the server 130 or a separate shopping mall management server (not shown) using the user terminal 150 . If a separate shopping mall management server exists, the shopping mall management server transmits the received product recommendation request to the server 130 .
  • the product recommendation request may include a query for similar products. Furthermore, it may include queries about similar customers. Then, when the processing of the query is completed using the graph-based data analysis method and/or the basket recommendation method, the processed data is provided to the user terminal 150 to confirm the purchase and select a product based on this make it possible
  • the user terminal 150 may provide a product recommendation request to the server 130 using the network 120 (a specific web page in the network).
  • the terminal or user terminal 150 associated with the shopping malls 110-1 to 110-N may download an application provided by the server 130 and automatically generate the application through the application. . That is, in the terminal or user terminal 150 associated with the shopping malls 110-1 to 110-N, when the user inputs a query for similar products or similar customers, a graph-based product recommendation process or product basket recommendation process is performed within the terminal. can be done in The processor (not shown) of the terminal executes the downloaded application without a separate server device, and it is possible to recommend an optimal product through data transformation and transformed data analysis.
  • FIG. 5 is a conceptual diagram for explaining a product recommendation method based on similarity.
  • the terms “device” or “product recommendation device” mentioned below may be interpreted as referring to the server 130 illustrated in FIG. 1 .
  • the device generally performs product recommendation based on the similarity of the product image and attributes. That is, first, by calculating the similarity between product images using the image analysis module, it can be used as a purchasability index.
  • the device may analyze the color and shape of the footwear product 1, and extract the footwear product 2 having the most similar color and shape from among numerous product images stored in the database.
  • the footwear product 2 having similar properties may be extracted. For example, if the color of footwear product 1 is red, the size is 250, the brand is “brand A”, and the shape is “shoes”, footwear product 2 with the same or similar attributes can be retrieved from the database and extracted. .
  • the method of using the degree of similarity based on the image and/or product attributes corresponds to the case of purchasing similar products at the same time, so it cannot be said that the probability of purchase is very high, and there is a case where it is not possible to meet the buyer's demand. Occurs.
  • the device may extract the attributes of products purchased a lot based on the purchase history of a specific customer, and set the purchaseability index of products having the corresponding attributes to be high. Using this point, it is possible to categorize the customer with the attributes of the product purchased by the customer to calculate the purchasability index. That is, it is possible to find customers who have a similar propensity to customers who have purchased the corresponding product, and recommend a product purchased by the similar propensity customer as a similar product.
  • the purchasability index based on various types of product images, product attributes, and/or customer attribute analysis can be used as an element to be inserted into the blank of a matrix in a product basket recommendation method later.
  • FIG. 6 is a graph based on product, order, and customer nodes used in a product recommendation method using a graph database according to an embodiment of the present invention.
  • the device may collect order history data to generate a graph composed of at least three node layers.
  • the order history data basically includes information that "a specific customer A ordered product C using order B at a specific shopping mall". That is, it is data including the relationship between customers, orders, and products.
  • the device generates a radial graph by forming products as a first node layer, orders as a second node layer, and customers as a third node layer, based on each order history data.
  • product A 310 is included in six orders 320 (orders A to F), and each order is obtained from a different customer 330 .
  • the device connects the order A 320 to the order F using a connecting line with the product A 310 as the center, and connects the customer A 330 who requested the order A 320 to the order A 320 .
  • the product-order connection line indicates the products included in the order
  • the order-customer connection line indicates the customer who placed an order.
  • the first node layer (product layer) of each node layer of the graph may be represented in red, the second node layer (order layer) in green, and the third node layer (customer layer) in blue.
  • the graph thus generated may be visualized and displayed through a display device.
  • the device may display a plurality of orders including one product 310 (product A) and a customer who placed the order.
  • product A product
  • customer A customer who placed the order.
  • FIG. 7 is a conceptual diagram for explaining a direct correlation and an indirect correlation between products.
  • the initial direct correlation (distance 2) (refer to the example of Fig. 7 (a)) shows a low degree of similarity, but when the distance increases by a certain interval (eg, distance 4), an indirect correlation can be observed.
  • This indirect correlation is preferably implemented as a high index in terms of purchaseability.
  • the number of paths through which two products (product 1 and product 2) can be connected at a predetermined distance (distance 4 in the embodiment of FIG. 7B ) may be an indicator of similarity.
  • customers connected by distance 4 may have a relationship of “other customers who purchased products purchased by a specific customer”.
  • customers may have a relationship of “other customers who purchased products purchased by a specific customer”.
  • the device may determine that the customer is the customer with the most connection paths, and that the customer B has the most similar purchasing propensity to the specific customer (customer A).
  • the device may recommend products that a specific customer A purchased but not purchased by a similar customer B to the customer B.
  • products purchased by customer B but not purchased by customer A may be recommended to A. That is, the device may connect customer A and customer B via the integrated product A 310 , and analyzes the similarity in consideration of the length of the connection path and the number of connection paths.
  • FIG. 8 is a flowchart illustrating a product recommendation method using a graph database according to an embodiment of the present invention.
  • a server collects order history data from an online shopping mall or a server managing the shopping mall (S510).
  • the acquired data includes information on a relationship between a product, an order including the product, and a customer who is the subject of the order.
  • the device converts the collected order history data into graph data (S512).
  • the graph is composed of three node hierarchies, and the product, which is the first node layer, is connected to the order, which is the second node layer, through a connection line, and the order is connected to the customer, which is the third node layer, through a connection line, to form the first node layer. It may have a graph form of a radial with a center. Then, the device stores the converted data in the form of a graph in the database (S514).
  • the client terminal that is, the user terminal inputs a recommended product search request (S516).
  • the input request information is transmitted to the device (server) side.
  • the recommended product search request may be divided into (i) a product-centered search request and (ii) a customer-centered search request. This will be described in more detail with reference to FIGS. 19 to 22 .
  • the device analyzes the correlation centered on the target product or target customer using the converted data in a graph form (S518). Then, with the target product or target customer as the center, a connection path to other products or customers is created based on the analyzed correlation (S520). At this time, based on the distance of the generated connection path, it is determined whether the connection path exists within a certain distance range (S522).
  • the number of connected paths increases proportionally, so that having an excessively large distance value is obtained.
  • It is preferable to exclude the connection path (eg, b 10).
  • the a and b values defining the distance range can be set through user settings and can be arbitrarily changed. In addition, values of a and b may be changed in a form of finding an optimal solution through machine learning. Connection paths having a distance less than a predetermined range or a distance exceeding the range are excluded (S524).
  • the device may adjust the number of connection paths in consideration of the distance of the connection paths between two nodes.
  • the distance between two directly connected nodes is defined as a distance of one hop.
  • the number of first connection paths can be calculated by adjusting 1/n1.
  • the number of second connection paths may be calculated by adjusting 1/n2. That is, except for the direct correlation of distance 2, since nodes connected by a short distance may have high similarity, the number of connection paths can be adjusted by dividing the nodes by the corresponding distance in consideration of this more importantly. For example, the number of connection paths between nodes at a distance of 4 is considered as 1/4, and may be considered larger (importantly) than 1/8, which is the number of connection paths at a distance of 8.
  • the device calculates the number of connection paths existing within the set distance range based on the above method (S526). Then, the product or customer having the largest number of connection paths is extracted (S528). In this case, by sorting based on the number of connection paths, it is possible to extract the top few products or customers as products or customers with high similarity.
  • a high degree of similarity means that the purchasing propensity is similar.
  • the target customer will purchase the remaining products excluding the integrated product in the first order including the integrated product through the integrated product. Since it can be determined that it is high, these products can be derived as recommended products.
  • the terminal may display the provided information to the purchaser by outputting it (S530).
  • FIG. 9 is a conceptual diagram illustrating a state in which a plurality of product-centered radial graphs are entangled.
  • a form in which a plurality of radial graphs 610 , 620 , 630 , and 640 are entangled may be represented.
  • the first radial graph 610 centered on product A and the second radial graph 620 centered on product B may be connected to each other via order A or customer A.
  • These two radial graphs 610 and 620 may be visualized and displayed on the display screen in two dimensions.
  • the first radial graph 610 and the third radial graph 630 centered on product C may be connected to each other via order B or customer B, and in this case, the second radial graph 620 and the third radial graph 630 .
  • Graph 630 may also be connected to each other via order C or customer C.
  • the device may display radial graphs in three dimensions. That is, if the first and second radial graphs 610 and 620 are displayed on the xy plane, the first and third radial graphs 610 and 630 may be displayed with a first angle in the z-axis direction, and the second The second and fourth radial graphs 620 and 640 may be displayed with an angle. Accordingly, graphs can be pulled through the zoom-in/zoom-out function of the user interface, and the three-dimensional angle can be adjusted using the panning function to view the connection between a plurality of radial graphs. can be visualized. And, when a specific node is selected, the selected node is brought to the center of the screen, so that a plurality of radial graphs connected to the central node can be displayed in all directions.
  • the type of relationship does not necessarily have to be taken as "Customer-Order-Product-Order-Customer", but the relationship of "Product-Order-Customer-Order-Product” is not necessarily taken as "Customer-Order-Product-Order-Customer” can be specified. In other words, it is okay to take various standards of relationship.
  • FIG. 10 is a diagram for explaining the appropriate use of the same/similar category and combination category in a product recommendation method using a graph database according to an embodiment of the present invention.
  • the device may utilize product category information when determining similarity, extracting similar products/customers, and deriving recommended products. For example, in finding similar products, similar products can be found only within the same category (eg, top), or in another category (as described above, categories by competition or combination) (eg, tops and bottoms, bottoms and shoes, etc.) ) can be found.
  • category information e.g., top, or in another category (as described above, categories by competition or combination) (eg, tops and bottoms, bottoms and shoes, etc.) ) can be found.
  • the first class is divided into categories of the largest category such as tops, bottoms, shoes, accessories, etc. can be subdivided and defined in Each product category is defined through an identifier, and by coding the similarity/combination relationship between the identifiers, it is possible to identify what kind of relationship each product category is.
  • the second class for example, it may be divided into a shirt, a blouse, a T-shirt, a hood, a vest, and the like, within the top.
  • the third class is defined as a subdivision of the second class.
  • similar products may be derived using the same/similar product category in all classes from the first class to the third class.
  • the "Southern” product category may be selected as a product category similar to "T-shirt” while being “top”, and a product belonging to the product category of "Southern” within a specific order is derived as a similar product.
  • similar products may be derived by using a combination category in the first class.
  • the “bottom” category is selected for “T-shirt”, and a product belonging to the bottom product category may be derived as a similar product of the T-shirt.
  • this setting may be appropriate, as it is often purchased in combination with a bottom.
  • the same/similar category may be used in the first class, but a combination category may be used in the second class.
  • a category of "top” may be selected, and in the second class, a "jumper” category combined with "T-shirt” may be selected. That is, a product belonging to the product category of “jumper” within a specific order may be derived as a similar product.
  • determining whether to use the same/similar product category or a combined product category is a factor that is considered important because the resulting values are different. Also, the result value is different depending on which class uses the similar category and the combined category.
  • the device may define these settings through user settings and allow the user to arbitrarily change them. Alternatively, big data-based artificial intelligence can be used to optimally adjust these areas. A method of utilizing the artificial intelligence model will be described in more detail with reference to FIG. 11 .
  • FIG. 11 is a conceptual diagram illustrating an AI learning method related to a recommendation algorithm used in a product recommendation method using a graph database according to an embodiment of the present invention.
  • the device learns the artificial intelligence model using the artificial intelligence learning module (not shown), it is possible to set whether to use a similar or combined category in a specific class based on the learned artificial intelligence model.
  • the artificial intelligence model used in the device includes a deep learning convolutional neural network model (CNN).
  • CNN deep learning convolutional neural network model
  • the present invention is not limited thereto, and may include any one of a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a Long Short Term Memory Network (LSTM).
  • DNN Deep Neural Network
  • RNN Recurrent Neural Network
  • LSTM Long Short Term Memory Network
  • the device generates order data 1 and order data 2 as training data sets.
  • the two order data are the previous order data (order data subject to the similarity search request) of the selection process according to the graph-based similar product derivation according to an embodiment of the present invention and the actual order after product recommendation. It has the relationship of order data. That is, the customer who ordered the order data 1 receives a plurality of product recommendations using the graph-based similar product derivation method according to an embodiment of the present invention, and combines the order data 2 made by selecting one of the recommended data. Create a training data set.
  • the optimal threshold value is found while accumulating the result value based on the training data set of order data 1 and order data 2 let it be
  • the node layer serving as a medium between the order data 1 and the order data 2 forming the learning data does not necessarily have to be the first node layer “product”, and may be the third node layer “customer”.
  • the second node layer may be a reference.
  • FIG. 12 is a conceptual diagram for explaining a method of indexing a connection path of a product recommendation method using a graph database according to an embodiment of the present invention.
  • the device performs indexing and identifies a connection path between a product and a product, a customer and a customer, or a product and a customer, in data.
  • the connection path may be composed of a node layer identifier and an index number.
  • the node layer identifier is composed of a combination of the identifier of the target node, which is the center of the search, and the identifier of the final destination node.
  • the first node layer (product node) may be identified as p
  • the third target node layer customer node
  • an indexing number is attached thereafter.
  • Indexing numbers are query-related and can have special meanings. For example, if the index "0100" is attached to the pp node layer identifier, this means that customers who have purchased the product are asking to find another product they have purchased, and the associated path is bundled by the customer who ordered the first product (departure node). other products ordered. A description of a query according to the meaning of an index directly following the node layer identifier will be described with reference to FIG. 13 .
  • each connection path is identified in the form of "-1" in the state where the index is attached. That is, the first path associated with the pp0100 query may be dataized as “pp0100-1”. After that, the number at the trailing end of "-" is incremented by 1 to identify the path. For example, if the customer who ordered the first product has a plurality of other products ordered in a bundle, it may be defined as pp0100-1, pp0100-2, pp0100-3, and the like.
  • connection path from the first customer to the second customer is expressed as mm0100-1, mm0100-2, or mm0101-1, mm0101-2, or the like.
  • a connection path from the first product to the first customer is expressed as pm0100-1, pm0100-2, and the like.
  • a connection path from the first customer to the first product is expressed as mp0100-1, mp0100-2, and the like.
  • connection paths are converted into data, and the device can retain them, and the number of connection paths within a specific distance range is calculated using the data.
  • the device may control the conversion from the implementation of the graph form of FIG. 9 to the form of a list of connection paths. That is, if there is a request for converting information related to a connection path between a specific departure node and a specific arrival node into a list form while visualizing a plurality of radial graphs, the connection path index between the departure node and the arrival node is converted into a list. can be displayed In addition, the distance of the connection path (eg, pp0100-1) of the first index may also be displayed in the corresponding list. Accordingly, when a connection path having a specific index is clicked, the corresponding path is displayed in a highlighted state on the graph, and it is visualized so that it can be intuitively seen which path connects from the departure node to the arrival node.
  • the connection path index between the departure node and the arrival node is converted into a list.
  • overlapping indexing may occur.
  • it is extracted as pp0101-1 so that it does not overlap, and when all paths are extracted, it can be identified as pp0101-1 expressing a more detailed connection path than that except for the overlapping pp0100-1.
  • 13 to 16 are diagrams illustrating various embodiments of finding similar products and customers with similar tendencies as outputs by designating product and customer nodes as inputs.
  • the device generates a query to find similar products and customers by specifying a product or customer node as input.
  • a query is generated to find a product having similarity to a specific product.
  • You can create a query pp0100 that shows other products that customers who ordered a product ordered in a batch.
  • a connection path can be derived by using the query pp0101 indicating a product in the same category among other products that the customer who ordered the product ordered in a bundle.
  • a query pp0200 indicating a product purchased by a customer who ordered a product in another order and a query pp0201 indicating a product in the same category in the other order may be used.
  • the following queries for finding customers who are similar to a specific customer You can use the mm0100 query that represents a second customer who ordered the same product as the first customer. Furthermore, the mm0101 query for finding a path to a second customer who ordered a second product of the same category rather than the exact same product as the first product ordered by the first customer may be used.
  • a query pm0000 indicating a path to a customer who purchased a product in a bundle with the corresponding product, and among the products purchased in a bundle, the corresponding product You can use the pm0100 query indicating the path of customers who purchased products in the same category as .
  • a query mp0000 indicating a path to bundled products such as a product ordered by the corresponding customer, and among the bundled purchased products, the corresponding You can use the query mp0001, which indicates a path to a product in the same category as the product.
  • a connection path to the query is also expressed as ppXXXX-X.
  • the connection path ppXXXX-X between the product and the product may be used, the combination of the pmXXXX-X path, which represents a path related to the similarity between the product and the customer, and the mpXXXX-X path, which represents a connection path related to the similarity between the customer and the product, may be used. is available. That is, the similarity between the product and the product may be determined using pm0000-1 + mp0000-1. Also, it is possible to determine the similarity between the product and the product by using three or more similarity-related connection paths, such as using pm0000-1 + mm0100-1 + mp0000-1.
  • 17 is a diagram illustrating a result of implementing a product recommendation method using a graph database in a web page according to an embodiment of the present invention.
  • the device executes the above algorithm by accessing it using a REST API.
  • the device designates which logic (relationship) to apply using a Uniform Resource Locator (URL).
  • URL Uniform Resource Locator
  • pp0100 is designated to derive other products purchased by members who have purchased the corresponding product.
  • a query parameter may be attached following a query such as pp0100, which may include a central product number, the number of data to be returned, and a member ID. That is, only customer or product information can be used according to the recommendation logic.
  • additional parameters such as product code and customer number are transmitted by URL encoding.
  • the query output will be provided with the number of data (n), products or members (nodes), importance (hits: number of routes searched) and cryptographic queries (Cypher queries for debugging purposes).
  • FIG. 18 is a flowchart illustrating a product recommendation method using a graph database according to another embodiment of the present invention.
  • the device collects inquiry history data related to product inquiry from shopping malls and/or a server managing the shopping malls ( S1210 ). This can be achieved by extracting the product information inquiry history of users from the web log file.
  • the inquiry history can be converted into sequential product codes and converted into data. For example, after viewing product 1, inquiring product 2, inquiring product 4, viewing product 1 again and logging out, taking into consideration the time-series part, the inquiry history data is generated by converting the product code into a product code.
  • the device converts the inquiry history data into a graph form (S1212) and stores it in the database (S1214).
  • the structure of the graph may have a structure of “session-page-product” to correspond to the previous “product-order-customer”.
  • the client may input a recommended product search request (S1216). As described with reference to FIG. 17 , this may be accomplished using a URL.
  • the device extracts the searched product within the range of x number of web pages after or before searching for a specific product (S1218).
  • the x value can be determined by user settings and can be arbitrarily changed.
  • the device may recommend to the purchaser in the order of the products having the highest number of views among the inquired products (S1220).
  • the client may receive and output the recommended product information (S1222).
  • 19 and 20 are diagrams illustrating exemplary aspects of product route prediction in a product recommendation method using a graph database according to another embodiment of the present invention.
  • the device when there is a recommended product request, the device generates a graph formed of at least three node layers. This may include a web page (1320: second node layer) searched by the user centering on the session 1330, which is the first node layer, and a product (1310: third node layer) searched for in the corresponding web page.
  • the node hierarchy may be changed so that the product becomes the first node hierarchy, the page becomes the second node hierarchy, and the session becomes the third node hierarchy.
  • the device connects the web page 1320 associated with the specific session 1330 , and configures the product 1310 searched for in the web page 1320 to be connected to the web page 1320 .
  • the device After composing the graph, the device searches for and recommends 10 products in the order of the number of hits within the range of 3 pages after the product 24522 (1310) inquiry. Contents related to the following three web pages will be described with reference to FIGS. 21 and 22 .
  • 21 and 22 are diagrams illustrating exemplary aspects of product route prediction in a situation in which the next product is inquired after the product inquiry in the embodiment of FIG. 19 .
  • the device uses the web page 1422 after the customer inquires for product 24522 (not shown) and then searches for product 25799 ( 1410 ) through the web page 1420 .
  • the fact that the product 25622 (1412) was inquired can be confirmed based on the converted graph data.
  • the client in FIGS. 19 to 20 retrieves and returns 10 products in the range of a specific number of pages (which may be 3 pages in this embodiment) in the order of the highest number of hits.
  • the three pages may be set to be the pages where the product search is made. That is, it is preferable to exclude web pages without product searches from calculating the number of strokes, count three pages based on pages on which product searches are made, and recommend top products that have been viewed a lot within the corresponding page range.
  • the learning data set may be a set of inquiry data 1, which is the target of a search request, and inquiry data 2, which is related to a product for which an actual purchase has been made through a graph-based product recommendation process according to an embodiment of the present invention.
  • the device may use a mixture of a first graph database based on product purchase history data and a second graph database based on product inquiry history data. For example, after the top 10 products with a large number of views are searched using the second graph database, the degree of similarity between the 10 recommended candidate products and the requested product may be calculated using the first graph database. Therefore, the recommendation order of the 10 recommended candidates may be changed based on products having a high similarity using the first graph database. The reverse is also possible.
  • weight 1 is given to the recommended product candidate calculated based on the first graph database
  • weight 2 is given to the recommended product candidate calculated based on the second graph database to finalize the product with the highest similarity based on the weight. Candidates may be considered.
  • 23 is a conceptual diagram for explaining a situation in which promotion of products whose sales are low according to product sales activity analysis is required.
  • the device performs an analysis of product sales volume and product sales activity, and promotes some of 10 products with low sales (products with activity lower than 0.2) (products in the lagger stage) This may be requested. That is, if you want to promote promotion by selecting 4 of 10 low-selling products, the product basket recommendation method according to an embodiment of the present invention may be used.
  • 24 and 25 are conceptual views for explaining a matching method for providing a special promotion by matching a specific number of customer groups and a specific number of product groups when there are 5 customers and 6 products.
  • the device may perform an optimized promotion in two ways.
  • the first method is a method of matching multiple product groups and multiple customer groups by first selecting the optimal product group for 5 people out of 6 products and secondarily selecting the optimal customer for the selected product (refer to FIG. 28) .
  • the second method is a method of matching multiple product groups and multiple customer groups by first selecting the optimal customer group for 6 products for 5 customers, and then selecting the optimal product for the selected customers (FIG. 32) Reference).
  • the device may use at least one of the above two methods to make a determination to optimize the sales potential, and by the determination, may proceed with the promotion in the state shown in FIG. 25 .
  • the device promotes a package consisting of product 2, product 3, and product 6 to customer 1 , customer 3 and customer 4 .
  • the promotion may be conducted in the form of providing a special discount on products included in the package to the customer only.
  • the product of the corresponding package may be displayed on the first page of the shopping mall in a pop-up method, or the package product may be arranged in a most conspicuous area to induce sales.
  • 26 and 27 are flowcharts illustrating a product basket recommendation method according to another embodiment of the present invention.
  • the device first acquires product and customer data that are candidates for promotion (S1710).
  • product and customer data that are candidates for promotion (S1710).
  • the aggregated data preferably takes the form of a matrix.
  • products are formed as rows and customers as columns.
  • an index related to the purchaseability of individual products and individual customers is input.
  • an index calculated based on product image analysis, product attribute analysis, and propensity according to customer purchase history may be used.
  • a similarity based on a graph database according to an embodiment of the present invention may be used as the indicator.
  • the similarity calculated by the number of connection paths using a graph database is used as the purchaseability index.
  • the device After acquiring data and generating a matrix, the device calculates the total number of connection paths for each product (S1720), and selects products with a higher rank with a large number of connection paths (S1730).
  • the number of selected products depends on the number of promotional products. This is a situation in which a user (eg, a shopping mall operator) can select through a user interface.
  • all products having a number of connection paths greater than or equal to a threshold value may be selected. For example, products having a connection path greater than the number of 700 connection paths may be selected without limiting the number.
  • a small number of connection paths may be judged to have low relevance. In this case, it is possible to select the one with a small number of connection paths and process it in the direction of increasing the discount.
  • the device selects the top four products with a high number of connection paths.
  • the device reconstructs the matrix with a new matrix by leaving only the top 4 selected products (which can be changed through user settings), and deleting the remaining products (S1740). Then, for the remaining 4 products, 6 customers with a high total number of connection paths are selected as target customers for the promotion. The device promotes the selected four product packages to six customers of the promotion target customers (S1750).
  • FIG. 28 is a flowchart illustrating a product-based basket recommendation method according to another embodiment of the present invention. This relates to the method (1) of FIG. 24 .
  • a matrix is generated in which the product is a column and the customer is a row ( S1810 ). In the blank of the matrix, the purchaseability index between the product and the customer is entered.
  • the device calculates the purchase possibility sum for each row (S1820). Then, the products are ranked in the order of the highest sum of purchasability, selected products having the highest order of purchasability by the number of products to be recommended by the promotion, and the remaining products are deleted to reconstruct the matrix (S1830).
  • the device calculates a subtotal for each customer for each column (S1840). Then, after ranking in order of increasing subtotals by customer, a customer having a high purchaseability subtotal rank as many as the number of customers to be promoted is selected and the remaining customers are excluded ( S1850 ).
  • the device When the selection of the customer group and the product group is completed, the device forms the product group as a package for the customer group and performs promotion (S1860).
  • 29 to 31 are tables for describing in detail a method of generating a matrix for each step of the basket recommendation method of FIG. 28 .
  • the apparatus generates a matrix by configuring customers as rows and products as columns based on the purchaseability indicators for customers 1 to 5 and products 1 to 6 .
  • the device ranks by calculating a subtotal for each product through the summation of rows. 30 , product 2 ranks first with a value of 69, product 3 ranks second with a value of 55, and product 6 ranks third with a value of 53 .
  • the top three products (product 2, product 3, and product 6) are selected, and a partial sum for each customer is calculated by summing the columns. Then, rank them.
  • customer 1 ranks first with a value of 48
  • customer 3 ranks second with a value of 43
  • customer 4 ranks third with a value of 33 .
  • the device selects customer 1, customer 3, and customer 4 as promotion target customers, and selects product 2, product 3, and product 6 as promotion target products to perform the promotion.
  • FIG. 32 is a flowchart illustrating a customer-based basket recommendation method according to another embodiment of the present invention. This relates to the method (2) of FIG. 24 .
  • a matrix is generated in which the product is a column and the customer is a row ( S2010 ). In the blank of the matrix, the purchaseability index between the product and the customer is entered.
  • the device calculates the purchasability sum for each column (S2020). Then, by ranking the customers in the order of the highest sum of purchasability, a customer having a high total purchasability rank is selected as much as the number of customers to be recommended, and the remaining customers are deleted to reconstruct the matrix (S2030).
  • the device calculates a partial sum for each product for each row (S2040). Then, after ranking in order of increasing subtotals by product, a product having a high purchasability subtotal rank by the number of products to be promoted is selected and the remaining products are excluded (S2050).
  • the device When the selection of the customer group and the product group is completed, the device forms the product group into a package for the customer group and performs promotion (S2060).
  • 33 to 35 are tables for describing in detail a method of generating a matrix for each step of the basket recommendation method of FIG. 32 .
  • the apparatus generates a matrix by configuring customers as rows and products as columns based on the purchaseability indicators for customers 1 to 5 and products 1 to 6 .
  • the device ranks by calculating subtotals for each customer by summing columns.
  • customer 1 ranks first with a value of 77
  • customer 3 ranks second with a value of 73
  • customer 4 ranks third with a value of 72 .
  • the top three customers (customer 1, customer 3, and customer 4) are selected, and the subtotals for each product are calculated by summing the rows. Then, rank them. 35 , product 2 ranks first with a value of 47, product 3 ranks second with a value of 41, and product 5 ranks third with a value of 37 .
  • the device selects customer 1, customer 3, and customer 4 as promotion target customers, and selects product 2, product 3, and product 5 as promotion target products to perform the promotion.
  • the promotional customer is the same but the product is different. As such, the results may be completely different depending on which method is used.
  • 36 is an apparatus for constructing customer big data based on generation of an integrated customer identifier according to an embodiment of the present invention, and performing a product recommendation method and/or a product basket recommendation method based on a graph database based on the constructed big data is a schematic block diagram of
  • the product recommendation apparatus may include a communication unit 2210 , a display unit 2220 , a processor 2230 , a memory 2240 , and an input unit 2250 . .
  • the communication unit 2210 is a component for communication with other devices by being connected to at least one of a wired network and a wireless network.
  • the communication unit 2210 may be implemented as an antenna, a communication-related chip, or the like.
  • the display unit 2220 is a component that displays information input/output through the communication unit 2210 and information processed through the processor 2230 . This may be implemented as one of a monitor, TV, and other panel.
  • the processor 2230 may receive a plurality of individual shopping mall customer identifiers corresponding to the plurality of shopping malls and a plurality of personal information element groups corresponding to the plurality of individual shopping mall customer identifiers from a plurality of shopping malls based on the communication unit.
  • the processor 2230 specifies a first customer based on the plurality of groups of personal information elements, generates an integrated customer identifier corresponding to the first customer, and sets the integrated customer identifier corresponding to the first customer.
  • Big data is stored in a database in association with a plurality of individual shopping mall customer identifiers and the plurality of personal information element groups, and by collecting log records of the first customer from the plurality of shopping malls in a database based on the integrated customer identifier can be built
  • the processor 2230 may be implemented as a microprocessor.
  • the processor 2230 may calculate a degree of similarity between a plurality of groups of personal information elements, and may specify the first customer based on the calculated degree of similarity.
  • the processor 2230 may calculate the number of personal information elements included in the personal information element group and the number of personal information elements that match between the personal information element groups. If the calculated similarity is greater than or equal to a predetermined value, the processor 2230 may specify the same customer.
  • the processor 2230 may input data of a plurality of personal information element groups into an artificial intelligence engine learned through deep learning and classify them by the same customer.
  • the processor 2230 receives, from a first shopping mall, a first individual shopping mall customer identifier capable of identifying a second customer in the first shopping mall and a first personal information element group corresponding to the first individual shopping mall customer identifier, and It is determined whether an integrated customer identifier corresponding to the second customer exists in the database based on the first individual shopping mall customer identifier, and if the integrated customer identifier corresponding to the second customer does not exist in the database, the Generate a unified customer identifier corresponding to the second customer based on the first personal information element group, and set the unified customer identifier corresponding to the second customer to the first individual shopping mall customer identifier and the first personal information element group It can be stored in the database in association with .
  • the processor 2230 is configured to compare the second customer's personal information element group and the first personal information element group stored in the database. Thus, it can be determined whether there is any change. In this case, if there is a changed content as a result of the comparison, the content of the second customer's personal information element group may be updated based on the changed content.
  • the processor 2230 performs a product recommendation corresponding to the first customer based on at least one of a product recommendation and a basket recommendation based on a graph database based on the collected log records of the first customer can do.
  • the memory 2230 may store data necessary for the operation of the processor 2230 .
  • the processor 2230 is a component that executes a graph-based similarity determination and product basket recommendation algorithm according to an embodiment of the present invention from another viewpoint.
  • the processor 2230 converts data in a graph form based on the order history data and inquiry history data input through the communication unit 2210, and uses the converted graph data in response to a query for a similar product from a client to return information about similar products and similar customers.
  • a similar product from a client to return information about similar products and similar customers.
  • the processor 2230 based on the number of connection paths from the first node to the second node (which may be limited to having a distance in a certain range), products and/or customers can be returned by sorting them in the order of the highest similarity. .
  • the processor 2230 may sort and return the products in the order of the most searched products in a page within a predetermined range before or after the first product inquiry.
  • the returned result data may be displayed through the display unit 2220 or provided to the requested terminal through the communication unit 2210 .
  • the processor 2230 generates a matrix in which the customer and the product are columns and rows, respectively, by obtaining the graph-based similarity information or other similarity index-based information on individual products and the purchaseability index of the individual customers. Then, by calculating a subtotal of rows and columns of the generated matrix, and ranking the calculated values, an optimal product package for a specific customer group can be generated, so that an appropriate promotion can be made.
  • the processor 2230 may include an artificial intelligence model learning module (not shown) and an artificial intelligence model execution module (not shown), based on the similarity of data of a plurality of personal information element groups.
  • an artificial intelligence model learning module not shown
  • an artificial intelligence model execution module not shown
  • the memory 2240 is a storage for storing instructions of a program executed by the processor 2230 .
  • the memory 2240 stores information received or transmitted through the communication unit 2210 .
  • at least a portion of information processed through the processor 2230 may be stored.
  • the input unit 2250 is a component that receives a user's input, and may be implemented as a keyboard, a mouse, a touch panel, or the like. Various parameters related to optimal product recommendation may be directly input through the input unit 2250 .
  • systems, devices, and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA). ), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose or special purpose computers.
  • the processing device may execute an operating system (OS) and one or more software applications running on the operating system.
  • a processing device may also access, store, manipulate, process, and generate data in response to execution of the software.
  • the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that can include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.
  • Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device.
  • the software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or apparatus, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave.
  • the software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.
  • the method according to the embodiments may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium.
  • the computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination.
  • the program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software.
  • Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks.
  • - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.
  • Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.
  • the hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Software Systems (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • General Business, Economics & Management (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un procédé, un dispositif et un système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré. Le procédé de production de mégadonnées de client peut comprendre les étapes suivantes : recevoir, d'une pluralité de centres commerciaux, une pluralité d'identifiants de client de centre commercial individuel correspondant à la pluralité de centres commerciaux et une pluralité de groupes d'éléments d'informations personnelles correspondant à la pluralité d'identifiants de client de centre commercial individuel ; spécifier un premier client en fonction de la pluralité de groupes d'éléments d'informations personnelles ; produire un identifiant de client intégré correspondant au premier client ; stocker, dans une base de données, l'identifiant de client intégré correspondant au premier client en association avec la pluralité d'identifiants de client de centre commercial individuel et la pluralité de groupes d'éléments d'informations personnelles ; et recueillir des enregistrements de journal du premier client auprès de la pluralité de centres commerciaux en fonction de l'identifiant de client intégré. Par conséquent, un document analysé par une analyse fiable dans les mégadonnées peut être utilisé pour un marketing efficace.
PCT/KR2021/008120 2020-06-30 2021-06-28 Procédé, dispositif et système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré WO2022005140A1 (fr)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
KR1020200079824A KR102538398B1 (ko) 2020-06-30 2020-06-30 빅데이터 기반의 그룹 단위의 상품 추천 방법, 장치 및 시스템
KR1020200079823A KR102518389B1 (ko) 2020-06-30 2020-06-30 고객 빅데이터를 활용한 상품 추천 방법, 장치 및 시스템
KR10-2020-0079822 2020-06-30
KR10-2020-0079823 2020-06-30
KR10-2020-0079824 2020-06-30
KR1020200079822A KR20220001616A (ko) 2020-06-30 2020-06-30 통합 고객 식별자 생성을 기반으로 하는 고객 빅데이터 구축 방법, 장치 및 시스템

Publications (1)

Publication Number Publication Date
WO2022005140A1 true WO2022005140A1 (fr) 2022-01-06

Family

ID=79315341

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2021/008120 WO2022005140A1 (fr) 2020-06-30 2021-06-28 Procédé, dispositif et système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré

Country Status (1)

Country Link
WO (1) WO2022005140A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006018340A (ja) * 2004-06-30 2006-01-19 Interscope Inc 顧客情報統合システム及び統合顧客情報データベースの作成法
KR20100038536A (ko) * 2008-10-06 2010-04-15 주식회사 이베이지마켓 인터넷을 이용한 전자상거래에서 고객정보 활용시스템 및 그 방법
KR20100095730A (ko) * 2009-02-23 2010-09-01 (주)유비전트 고객 정보 통합 시스템 및 방법
KR20150121945A (ko) * 2014-04-22 2015-10-30 주식회사 포워드벤처스 아이템 추천 시스템 및 아이템 추천 방법
KR20200052448A (ko) * 2018-10-30 2020-05-15 삼성전자주식회사 지식 그래프에 기초하여 데이터베이스들을 통합하는 시스템 및 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006018340A (ja) * 2004-06-30 2006-01-19 Interscope Inc 顧客情報統合システム及び統合顧客情報データベースの作成法
KR20100038536A (ko) * 2008-10-06 2010-04-15 주식회사 이베이지마켓 인터넷을 이용한 전자상거래에서 고객정보 활용시스템 및 그 방법
KR20100095730A (ko) * 2009-02-23 2010-09-01 (주)유비전트 고객 정보 통합 시스템 및 방법
KR20150121945A (ko) * 2014-04-22 2015-10-30 주식회사 포워드벤처스 아이템 추천 시스템 및 아이템 추천 방법
KR20200052448A (ko) * 2018-10-30 2020-05-15 삼성전자주식회사 지식 그래프에 기초하여 데이터베이스들을 통합하는 시스템 및 방법

Similar Documents

Publication Publication Date Title
WO2019245316A1 (fr) Système et procédé de génération de recommandations sur la base de descriptions explicables d'aspect amélioré
WO2020222623A2 (fr) Système et procédé pour construire automatiquement un contenu destiné à des ventes stratégiques
WO2020171535A2 (fr) Procédé pour fournir un service de recommandation d'article de mode en utilisant la forme du corps et l'historique d'achats de l'utilisateur
WO2010119996A1 (fr) Procédé et dispositif de fourniture d'annonces publicitaires à images mobiles
WO2016013914A1 (fr) Procédé, appareil, système et programme d'ordinateur permettant de fournir et d'afficher des informations de produit
WO2012091360A2 (fr) Procédé et système de fourniture de contenu personnalisé par l'utilisateur
KR20200088131A (ko) 그래프 데이터베이스를 이용한 상품 추천 방법 및 장치
WO2023022284A1 (fr) Plateforme de médiation de matériau d'emballage et procédé
WO2014035048A1 (fr) Système et procédé de recommandation d'articles
WO2017026852A1 (fr) Procédé pour un service de recommandation de produit par l'intermédiaire d'une comparaison relative, appareil et programme de service de recommandation associés
WO2020166765A1 (fr) Système de gestion d'habitude alimentaire et procédé d'utilisation associé
WO2017116215A1 (fr) Système et procédé de sélection de projet de recherche et de développement par des propositions autonomes d'indicateurs d'évaluation
KR102518389B1 (ko) 고객 빅데이터를 활용한 상품 추천 방법, 장치 및 시스템
WO2020022819A1 (fr) Communication par le biais d'un utilisateur simulé
KR102538398B1 (ko) 빅데이터 기반의 그룹 단위의 상품 추천 방법, 장치 및 시스템
WO2024025220A1 (fr) Système pour fournir une plateforme de contenu publicitaire en ligne
KR102193345B1 (ko) 그룹 고객과 그룹 상품 간의 상품 추천 방법 및 장치
WO2018043859A1 (fr) Dispositif d'analyse automatique de préférence d'article de location au moyen d'une image d'utilisateur et procédé correspondant
WO2013032198A1 (fr) Moteur de recommandation basé sur des articles pour recommander un article fortement associé
WO2010032918A9 (fr) Système et procédé de gestion d'informations d'affichage de contenu
WO2022005140A1 (fr) Procédé, dispositif et système de construction de mégadonnées de client basée sur la production d'un identifiant de client intégré
US20060122911A1 (en) E-commerce system and method
WO2024005242A1 (fr) Procédé de traitement de données concernant une transaction pour œuvre d'art
WO2022154213A1 (fr) Procédé et dispositif de génération de liste de correspondance utilisateurs-annonces d'annonce en ligne
WO2022244905A1 (fr) Procédé de gestion d'informations de flux d'articles et appareil associé

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21832023

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21832023

Country of ref document: EP

Kind code of ref document: A1