Member data selection method and system based on artificial intelligence

Abstract

The invention is suitable for the field of data screening, and provides a member data selection method and system based on artificial intelligence, and the method comprises the steps: obtaining and storing intra-domain user data from a plurality of target data sources, and the target data sources comprise but are not limited to an online sales platform, an offline store system, an intra-domain user management system and a social media platform of a target enterprise; establishing an intra-domain user information set database based on the intra-domain user data obtained by the plurality of target data sources; extracting target feature data in the intra-domain user information set database; presetting a circle selection rule and a training model, and respectively generating users in an intention purchase domain, users in a high-value domain and users in a potential loss domain; based on the users in the intention purchase domain, the users in the high-value domain and the users in the potential loss domain, generating associated reaction actions; and obtaining and verifying a circling result, and generating feedback information, so that a comprehensive intra-domain user portrait can be formed, deep understanding of intra-domain user demands and behaviors is facilitated, multi-dimensional targeted feature data is extracted, different types of intra-domain users are accurately circled, and personalized strategies are formulated for different intra-domain user groups.

Description

Technical Field

[0001] The present invention belongs to the field of data screening, and in particular to a member data selection method and system based on artificial intelligence. Background Art

[0002] In today's business environment, membership system has become an important means for enterprises to attract and retain customers. Through the membership system, enterprises can collect a large amount of information about members, including basic information, consumption behavior, preferences, etc. These member data contain rich value and are of vital importance to enterprises in precision marketing, personalized services, customer relationship management and market analysis.

[0003] Currently, member data selection is mostly done manually. Manual selection of member data is easily affected by subjective factors, resulting in unstable accuracy of the selection results. Different operators may have different understandings and judgments of the data, resulting in deviations when setting screening conditions. With the rapid development of artificial intelligence technology, especially the continuous breakthroughs in machine learning, deep learning and other fields, new ideas and possibilities have been provided to solve the difficulties of traditional member data selection methods. In order to improve the reliability of member data selection, it is urgent to develop a member data selection method and system based on artificial intelligence. Summary of the invention

[0004] The purpose of the present invention is to provide a member data selection method and system based on artificial intelligence, aiming to solve the problems raised in the above background technology.

[0005] The present invention is implemented in this way. On the one hand, a member data selection method based on artificial intelligence comprises: Acquire and store domain user data from several target data sources, including but not limited to the target enterprise's online sales platform, offline store system, domain user management system, and social media platform; Based on the in-domain user data obtained from several target data sources, a user information collection database in the domain is established; Extracting targeted feature data from a user information collection database within the domain; Preset selection rules and training models to generate users in the domain with the intention to purchase, users in the domain with high value, and users in the domain with potential churn; Generate related reaction actions based on users in the purchase intention domain, high-value domain users and potential churn domain users; Obtain and verify the circled results and generate feedback information.

[0006] As a further solution of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively specifically include: Obtain targeted products; Based on the targeted feature data, extract the in-domain user information whose number of targeted product page traffic exceeds the number threshold within the target period, and generate a first sub-domain user set; Based on the targeted feature data, extract the information of users in the domain who added the targeted product to the shopping cart but did not pay within the payment period threshold, and generate a second sub-domain user set; Identify and extract the repeated intra-domain user information in the first sub-domain user set and the second sub-domain user set, generate the intra-domain users who intend to purchase, and obtain the contact information of the intra-domain users who intend to purchase.

[0007] As a further solution of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively further include: Based on the targeted feature data, extract the information of users in the domain whose cumulative consumption amount reaches the consumption amount threshold, and generate a user set in the third subdomain; Based on the targeted feature data, extracting information of users in the domain whose consumption frequency reaches a first consumption frequency threshold, and generating a fourth sub-domain user set; Identify and extract repeated intra-domain user information in the third sub-domain user set and the fourth sub-domain user set, generate high-value intra-domain users, and obtain high-value intra-domain user contact information.

[0008] As a further solution of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively further include: Based on the targeted feature data, extract the in-domain user information that unfollows the target enterprise on the online sales platform and social media platform to generate the fifth in-domain user set; Based on the targeted feature data, extracting the in-domain user information below the second consumption frequency threshold, and generating a sixth in-domain user set; Identify and extract the repeated intra-domain user information in the fifth sub-domain user set and the sixth sub-domain user set, generate potential lost intra-domain users, and obtain the potential lost intra-domain user contact information.

[0009] As a further solution of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Based on the contact information of users in the intended purchase domain, obtain and push targeted product discount information and positive evaluation information to users in the intended purchase domain; Regularly collect purchase information of targeted products of users in the intended purchase domain.

[0010] As a further solution of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Generate and push exclusive discount information based on high-value domain user contact information; Regularly collect consumption information of users in high-value domains.

[0011] As a further solution of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Generate and push return gift discount information based on the contact information of users in the potential churn domain; Regularly collect consumption and payment information of users in the potential churn domain.

[0012] As a further solution of the present invention, on the other hand, a member data selection system based on artificial intelligence, the system comprises: A first acquisition module, used to acquire and store user data in the domain from a plurality of target data sources; A database module is used to establish a user information collection database within the domain based on the user data within the domain obtained from several target data sources; An extraction module, used to extract targeted feature data from a user information collection database within a domain; The preset module is used to preset selection rules and train models; The first generation module is used to generate users in the domain with the intention to purchase, users in the domain with high value, and users in the domain with potential churn; The second generation module is used to generate associated reaction actions based on the users in the domain with the intention to purchase, the users in the domain with high value, and the users in the domain with potential churn; The second acquisition module is used to acquire and verify the selection result; The third generating module is used to generate feedback information.

[0013] The present invention provides an artificial intelligence-based member data selection method and system. The method and system facilitate the formation of a comprehensive domain user portrait by integrating online and offline multi-channel domain user data, help to deeply understand the needs and behaviors of domain users, extract multi-dimensional targeted feature data, accurately select different types of domain users, and formulate personalized strategies for different domain user groups, thereby improving marketing accuracy, enhancing the loyalty of high-value domain users, reducing churn rate, and improving data selection reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a main flow chart of the member data selection method based on artificial intelligence.

[0015] Figure 2 The present invention is a flowchart of the first embodiment of a member data selection method based on artificial intelligence, in which selection rules are preset to respectively generate users in a domain with the intention to purchase, users in a domain with high value, and users in a domain with potential churn.

[0016] Figure 3 The present invention is a flowchart of a first implementation scheme of a member data selection method based on artificial intelligence, in which selection rules are preset to respectively generate users in a domain with the intention to purchase, users in a domain with high value, and users in a domain with potential churn.

[0017] Figure 4 The present invention is a flowchart of a third embodiment of a member data selection method based on artificial intelligence, in which selection rules are preset to respectively generate users in a domain with the intention to purchase, users in a domain with high value, and users in a domain with potential churn.

[0018] Figure 5 The present invention is a flow chart of the first embodiment of a member data selection method based on artificial intelligence, in which selection rules are preset to generate users in the domain with the intention to purchase, users in the domain with high value and users in the domain with potential loss.

[0019] Figure 6 The invention is a flow chart of a second embodiment of a member data selection method based on artificial intelligence, in which selection rules are preset to generate users in a domain with the intention to purchase, users in a domain with high value and users in a domain with potential loss.

[0020] Figure 7 The present invention is a flowchart of a third embodiment of a member data selection method based on artificial intelligence, in which selection rules are preset to generate users in a domain with the intention to purchase, users in a domain with high value, and users in a domain with potential loss.

[0021] Figure 8 It is a main structure diagram of a member data selection system based on artificial intelligence. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0023] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0024] The present invention provides an artificial intelligence-based member data selection method and system, which solves the technical problems in the background technology.

[0025] like Figure 1FIG. 1 is a main flow chart of a member data selection method based on artificial intelligence provided by an embodiment of the present invention. The member data selection method based on artificial intelligence includes: Step S100: Acquire and store domain user data from a number of target data sources, the target data sources including but not limited to the target enterprise's online sales platform, offline store system, domain user management system, and social media platform; Step S200: establishing a user information collection database within the domain based on the user data within the domain obtained from a number of target data sources; Step S300: extracting targeted feature data from the intra-domain user information collection database; Step S400: Preset selection rules and training models to generate users in the domain with the intention to purchase, users in the domain with high value, and users in the domain with potential churn; Step S500: generating a related reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value, and the users in the domain with potential churn; Step S600: Obtain and verify the circle selection result and generate feedback information; When this embodiment is applied, the in-domain user is a member customer of the target enterprise. The target enterprise is equipped with multiple sales channels and an in-domain user management system. The multiple sales channels include online sales platforms, offline store systems, and social media platforms. In the process of storing in-domain user data, the acquired in-domain user data should also be cleaned and preprocessed, and then an in-domain user information collection database is established based on the in-domain user data acquired from several target data sources to facilitate subsequent data analysis and processing, and then targeted feature data in the in-domain user information collection database is extracted, and the targeted feature data includes consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics. The consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics all contain a variety of information data of this category, and preset selection rules and training models are used. Through the selection rules and training models, based on the consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics, the required in-domain user information is selected through algorithm calculation and pattern recognition of the training model, and then the intended purchase in-domain users, high-value in-domain users, and potential loss in-domain users are generated respectively. For the intended purchase in-domain users, the goal is to accurately identify Identify the user groups in the domain with a higher possibility of purchase, so as to push promotional information or provide purchase guidance in a targeted manner. For high-value domain users, the goal is to identify the user groups in the domain that contribute more to the enterprise, provide them with exclusive services and discounts, and enhance their loyalty and consumption stickiness. For potential churn domain users, the goal is to identify the domain users who may churn in advance and take recovery measures. After selecting the domain users with intention to purchase, high-value domain users and potential churn domain users, based on the domain users with intention to purchase, high-value domain users and potential churn domain users, Generate associated reaction actions, obtain and verify the selection results after a certain period, generate feedback information, and evaluate the accuracy by comparing the actual business data and the selection results; this method and system integrates online and offline multi-channel domain user data to facilitate the formation of a comprehensive domain user portrait, help to deeply understand the needs and behaviors of domain users, extract multi-dimensional targeted feature data, accurately select different types of domain users, and formulate personalized strategies for different domain user groups, thereby improving marketing accuracy, enhancing the loyalty of high-value domain users, reducing churn rate, and improving domain user satisfaction.

[0026] like Figure 2 As shown, as a preferred embodiment of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively specifically include: Step S401: Obtain targeted products; Step S402: extracting information of users in the domain whose traffic times on the targeted product page exceed the threshold in the target period based on the targeted feature data, and generating a first sub-domain user set; Step S403: Based on the targeted feature data, extract the information of users in the domain who added the targeted product to the shopping cart but did not pay within the payment period threshold, and generate a second sub-domain user set; Step S404: identifying and extracting repeated intra-domain user information from the first sub-domain user set and the second sub-domain user set, generating potential purchase intra-domain users, and obtaining potential purchase intra-domain user contact information; When this embodiment is applied, a number threshold and a payment period threshold are preset. The information of in-domain users whose number of targeted product page traffic exceeds the number threshold within the target period and the information of in-domain users who add the targeted product to the shopping cart but do not pay within the payment period threshold belong to the preference behavior characteristics. First, the targeted product is obtained. The targeted feature data in the in-domain user information set database includes consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics. The consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics all contain a variety of information data of this category. Based on the targeted feature data, the information of in-domain users whose number of targeted product page traffic exceeds the number threshold within the target period is extracted to generate a first sub-domain user set. Based on the targeted feature data, the information of in-domain users who add the targeted product to the shopping cart but do not pay within the payment period threshold is extracted to generate a second sub-domain user set. The repeated in-domain user information in the first sub-domain user set and the second sub-domain user set is identified and extracted to generate users in the domain who intend to purchase, and the contact information of the users in the domain who intend to purchase is obtained.

[0027] like Figure 3 As shown, as a preferred embodiment of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively further include: Step S411: extracting information of users in the domain whose cumulative consumption amount reaches the consumption amount threshold based on the targeted feature data, and generating a user set in the third subdomain; Step S412: extracting information of users in the domain whose consumption frequency reaches a first consumption frequency threshold based on the targeted feature data, and generating a fourth sub-domain user set; Step S413: identifying and extracting repeated intra-domain user information in the third sub-domain user set and the fourth sub-domain user set, generating high-value intra-domain users, and obtaining high-value intra-domain user contact information; When this embodiment is applied, a consumption amount threshold and a first consumption frequency threshold are preset, and the in-domain user information whose cumulative consumption amount reaches the consumption amount threshold and the in-domain user information whose consumption frequency reaches the first consumption frequency threshold are termed consumption behavior features. Based on the targeted feature data, the in-domain user information whose cumulative consumption amount reaches the threshold amount is extracted to generate a third sub-domain user set. Based on the targeted feature data, the in-domain user information whose consumption frequency reaches the frequency threshold is extracted to generate a fourth sub-domain user set. The repeated in-domain user information in the third sub-domain user set and the fourth sub-domain user set is identified and extracted to generate high-value in-domain users, and the contact information of the high-value in-domain users is obtained.

[0028] like Figure 4 As shown, as a preferred embodiment of the present invention, the preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively further include: Step S421: extracting information of in-domain users who have unfollowed the target enterprise from the online sales platform and the social media platform based on the targeted feature data, and generating a fifth in-domain user set; Step S422: extracting the intra-domain user information below the second consumption frequency threshold based on the targeted feature data, and generating a sixth intra-domain user set; Step S423: identifying and extracting repeated intra-domain user information in the fifth sub-domain user set and the sixth sub-domain user set, generating potential lost intra-domain users, and obtaining contact information of potential lost intra-domain users; It should be understood that a second consumption frequency threshold is preset, and the domain user information of unfollowing the target enterprise in the online sales platform and the social media platform and the domain user information below the second consumption frequency threshold belong to consumption behavior characteristics and social behavior characteristics. Based on the targeted feature data, the domain user information of unfollowing the target enterprise in the online sales platform and the social media platform is extracted to generate a fifth domain user set. Based on the targeted feature data, the domain user information below the second consumption frequency threshold is extracted to generate a sixth domain user set. The repeated domain user information in the fifth sub-domain user set and the sixth sub-domain user set is identified and extracted to generate potential lost domain users, and the contact information of the potential lost domain users is obtained.

[0029] like Figure 5 As shown, as a preferred embodiment of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Step S501: based on the contact information of users in the intended purchase domain, obtaining and pushing targeted commodity discount information and positive evaluation information of users in the intended purchase domain; Step S502: regularly collecting purchase information of targeted commodities of users in the intended purchase domain; When this embodiment is applied, first, based on the contact information of users in the intended purchasing domain, discount information and positive evaluation information of targeted commodities for users in the intended purchasing domain are obtained and pushed, promotional information is pushed or purchase guidance is provided in a targeted manner, and purchase information of targeted commodities for users in the intended purchasing domain is collected at regular intervals, so as to compare the accuracy of actual business data and selection results through purchase information (whether an order is placed).

[0030] like Figure 6 As shown, as a preferred embodiment of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Step S511: Generate and push exclusive discount information based on the contact information of users in the high-value domain; Step S512: regularly collecting consumption amount information of users in the high-value domain; When this embodiment is applied, it first generates and pushes exclusive discount information based on the contact information of users in the high-value domain to enhance their loyalty and consumption stickiness, and regularly collects consumption amount information of users in the high-value domain to facilitate comparison of actual business data and accuracy of circled results through consumption amount (whether the total consumption amount has increased).

[0031] like Figure 7 As shown, as a preferred embodiment of the present invention, the generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Step S521: Generate and push return gift discount information based on the contact information of potential churn domain users; Step S522: regularly collecting consumption payment information of users in the potential churn domain; When this embodiment is applied, it first generates and pushes return gift discount information based on the contact information of users in the potential churn domain, improves the user consumption experience and the degree of care, and regularly collects the consumption payment information of users in the potential churn domain, so as to facilitate the comparison of actual business data and the accuracy of the circled results through the consumption payment information (whether they consume again).

[0032] like Figure 8 As shown, as another preferred embodiment of the present invention, on the other hand, a member data selection system based on artificial intelligence, the system comprises: The first acquisition module 100 is used to acquire and store user data in the domain from a plurality of target data sources; The database module 200 is used to establish a user information collection database in the domain based on the user data in the domain obtained from several target data sources; Extraction module 300, used to extract targeted feature data from the intra-domain user information set database; A preset module 400 is used to preset selection rules and training models; The first generating module 500 is used to generate the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss respectively; The second generating module 600 is used to generate associated reaction actions based on the users in the domain with the intention to purchase, the users in the domain with high value, and the users in the domain with potential churn; The second acquisition module 700 is used to acquire and verify the circle selection result; The third generating module 800 is used to generate feedback information; The first acquisition module 100 acquires and stores domain user data from several target data sources. Based on the domain user data acquired from the several target data sources, the database module 200 establishes a domain user information set database. The extraction module 300 extracts targeted feature data from the domain user information set database. The preset module 400 presets selection rules and training models. The first generation module 500 generates users in the domain with the intention to purchase, high-value users in the domain and potential churn users in the domain respectively. Based on the users in the domain with the intention to purchase, high-value users in the domain and potential churn users in the domain, the second generation module 600 generates associated reaction actions. The second acquisition module 700 acquires and verifies the selection results. The third generation module 800 generates feedback information.

[0033] In the above embodiment of the present invention, a member data selection method based on artificial intelligence is provided, and a member data selection system based on artificial intelligence is provided. The target enterprise is provided with multiple sales channels and an intra-domain user management system. The multiple sales channels include online sales platforms, offline store systems, and social media platforms. In the process of storing intra-domain user data, the acquired intra-domain user data should also be cleaned and pre-processed, and then an intra-domain user information collection database is established based on the intra-domain user data acquired from several target data sources for subsequent data analysis and processing, and then targeted feature data in the intra-domain user information collection database is extracted, the targeted feature data includes consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics, and the consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics all contain a variety of information data of this category, and selection rules and training models are preset. Through the selection rules and training models, based on the consumption behavior characteristics, preference behavior characteristics, and social behavior characteristics, the required intra-domain user information is selected through algorithm calculation and pattern recognition of the training model, and then the intended purchase intra-domain users, high-value intra-domain users, and potential loss intra-domain users are generated respectively. For the intended purchase For in-domain users who intend to buy, the goal is to accurately identify in-domain user groups with a high possibility of purchase, so as to push promotional information or provide purchase guidance in a targeted manner. For high-value in-domain users, the goal is to identify in-domain user groups that contribute more to the company, provide them with exclusive services and discounts, and enhance their loyalty and consumption stickiness. For potential churn in-domain users, the goal is to identify in-domain users who may churn in advance and take recovery measures. After selecting in-domain users with intention to buy, high-value in-domain users and potential churn in-domain users, based on in-domain users with intention to buy, high-value in-domain users and potential churn, For lost in-domain users, associated reaction actions are generated. After a certain period, the selection results are obtained and verified, feedback information is generated, and accuracy is evaluated by comparing actual business data with the selection results. This method and system facilitate the formation of a comprehensive in-domain user portrait by integrating online and offline multi-channel in-domain user data, helping to gain an in-depth understanding of the needs and behaviors of in-domain users, extract multi-dimensional targeted feature data, accurately select different types of in-domain users, and formulate personalized strategies for different in-domain user groups, thereby improving marketing accuracy, enhancing the loyalty of high-value in-domain users, reducing the churn rate, and improving the reliability of data selection.

[0034] In order to load the above-mentioned method and system and enable it to run smoothly, the system, in addition to the various modules mentioned above, may also include more or fewer components than described above, or a combination of certain components, or different components, for example, it may include input and output devices, network access devices, buses, processors and memories, etc.

[0035] The processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the system, and various interfaces and lines are used to connect various parts.

[0036] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0037] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0038] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A member data selection method based on artificial intelligence, It is characterized in that The method comprises: Acquire and store domain user data from several target data sources, including but not limited to the target enterprise's online sales platform, offline store system, domain user management system, and social media platform; Based on the in-domain user data obtained from several target data sources, a user information collection database in the domain is established; Extracting targeted feature data from a user information collection database within the domain; Preset selection rules and training models to generate users in the domain with the intention to purchase, users in the domain with high value, and users in the domain with potential churn; Generate related reaction actions based on users in the purchase intention domain, high-value domain users and potential churn domain users; Obtain and verify the circled results and generate feedback information.

2. According to the member data selection method based on artificial intelligence according to claim 1, It is characterized in that The preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively specifically include: Obtain targeted products; Based on the targeted feature data, extract the in-domain user information whose number of targeted product page traffic exceeds the number threshold within the target period, and generate a first sub-domain user set; Based on the targeted feature data, extract the information of users in the domain who added the targeted product to the shopping cart but did not pay within the payment period threshold, and generate a second sub-domain user set; Identify and extract the repeated intra-domain user information in the first sub-domain user set and the second sub-domain user set, generate the intra-domain users who intend to purchase, and obtain the contact information of the intra-domain users who intend to purchase.

3. According to the member data selection method based on artificial intelligence according to claim 1, It is characterized in that The preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively specifically include: Based on the targeted feature data, extract the information of users in the domain whose cumulative consumption amount reaches the consumption amount threshold, and generate a user set in the third subdomain; Based on the targeted feature data, extracting information of users in the domain whose consumption frequency reaches a first consumption frequency threshold, and generating a fourth sub-domain user set; Identify and extract repeated intra-domain user information in the third sub-domain user set and the fourth sub-domain user set, generate high-value intra-domain users, and obtain high-value intra-domain user contact information.

4. According to the member data selection method based on artificial intelligence according to claim 1, It is characterized in that The preset selection rules for generating the intended purchase domain users, high-value domain users and potential churn domain users respectively specifically include: Based on the targeted feature data, extract the in-domain user information that unfollows the target enterprise on the online sales platform and social media platform to generate the fifth in-domain user set; Based on the targeted feature data, extracting the in-domain user information below the second consumption frequency threshold, and generating a sixth in-domain user set; Identify and extract the repeated intra-domain user information in the fifth sub-domain user set and the sixth sub-domain user set, generate potential lost intra-domain users, and obtain the potential lost intra-domain user contact information.

5. According to the member data selection method based on artificial intelligence according to claim 2, It is characterized in that The generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Based on the contact information of users in the intended purchase domain, obtain and push targeted product discount information and positive evaluation information to users in the intended purchase domain; Regularly collect purchase information of targeted products of users in the intended purchase domain.

6. The member data selection method based on artificial intelligence according to claim 3, It is characterized in that The generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Generate and push exclusive discount information based on high-value domain user contact information; Regularly collect consumption information of users in high-value domains.

7. According to the member data selection method based on artificial intelligence according to claim 4, It is characterized in that The generating of the associated reaction action based on the users in the domain with the intention to purchase, the users in the domain with high value and the users in the domain with potential loss specifically includes: Generate and push return gift discount information based on the contact information of users in the potential churn domain; Regularly collect consumption and payment information of users in the potential churn domain.

8. An artificial intelligence-based intra-domain user data selection system, It is characterized in that The member data selection method based on artificial intelligence as described in any one of claims 1 to 7 is applied, and the system comprises: A first acquisition module, used to acquire and store user data in the domain from a plurality of target data sources; A database module is used to establish a user information collection database within the domain based on the user data within the domain obtained from several target data sources; An extraction module, used to extract targeted feature data from a user information collection database within a domain; The preset module is used to preset selection rules and train models; The first generation module is used to generate users in the domain with the intention to purchase, users in the domain with high value, and users in the domain with potential churn; The second generation module is used to generate associated reaction actions based on the users in the domain with the intention to purchase, the users in the domain with high value, and the users in the domain with potential churn; The second acquisition module is used to acquire and verify the selection result; The third generating module is used to generate feedback information.

Images