A customer relationship management system and method thereof

Abstract

The present invention discloses a customer relationship management system and a method thereof, comprising a server module, a user information module, a database module, a storage module, and a communication module. The server module is used for architecture services between the modules to facilitate the management of customer relationships; the user information module is used for collecting and screening basic information of customers; the database module is used for establishing an information base based on customer resource data and store information data; the communication module is used for communication in preliminary consultation, sales communication, and after-sales service. The present invention can facilitate the understanding of user information, and can facilitate the classification of users in conjunction with the database module, facilitate the push of suitable commodities, and facilitate users to purchase and understand commodity information. By effectively managing customer information and needs, customer satisfaction and loyalty can be improved, which not only helps to retain old customers, but also can attract new customers in the fierce market competition.

Description

Technical Field

[0001] The present invention relates to the technical field of customer relationship management, and in particular to a customer relationship management system and a method thereof. Background Art

[0002] Customer relationship management system, referred to as CRM, is essentially a computer management software that can store and classify the user's customer information, and the customer information can be easily retrieved. Due to the electronic data characteristics of computer software, it brings many significant conveniences to users and is rapidly being promoted and applied.

[0003] The customer relationship management systems currently on the market mainly focus on managing basic customer information and communication records, and the data processed is relatively simple. These systems are difficult to fully reflect the complex personnel relationships within customers, nor can they effectively record the connection between customers and their related implementation projects. They cannot classify users according to their information, etc., and cannot push communications quickly and accurately. Therefore, it is necessary to propose a customer relationship management system and its method to solve the above-mentioned problems. Summary of the invention

[0004] The object of the present invention is to provide a customer relationship management system and method thereof to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A customer relationship management system, comprising a server module, a user information module, a database module, a storage module, and a communication module, wherein the server module is used for architecture services between the modules to facilitate the management of customer relationships;

[0006] The user information module is used to collect and filter the basic information of customers;

[0007] The database module is used to establish an information base based on customer resource data and store information data. The store information data includes the classification and summary of commodities. The commodities are classified into primary, secondary and tertiary levels.

[0008] The storage module is used to store user information, transaction records, communication information, store information, and information databases. It cooperates with the database module to store the classification and summary information of commodities and the correlation degree of customer resource data in different files, so as to facilitate subsequent users to browse and push information when purchasing;

[0009] The customer resource data relevance classification and storage includes the following steps:

[0010] S31: extract all relevant features or indicators from customer resource data, perform dimensionless processing, and eliminate different dimensions and orders of magnitude of different features;

[0011] S32: Use grey correlation analysis to calculate the correlation between each feature, establish a correlation matrix, and evaluate the similarity between each feature by calculating the correlation coefficient;

[0012] S33: Classify the features according to the degree of association, set several levels, and assign the features to corresponding levels according to the degree of association. When the user browses or purchases the products subsequently, the products are pushed according to the degree of association.

[0013] The communication module is used for preliminary consultation, sales communication and after-sales service.

[0014] Preferably, the user information module includes personal identity information and network activity information; personal identity information mainly includes the user's name, date of birth, ID number, address, telephone number, account number and password, preferences; network activity information includes the user's IP address, Cookie information, online and technical information such as Internet or other electronic network activity information.

[0015] Preferably, the database module includes a customer resource data module and a store information data module. The customer resource module includes social characteristic data, purchase behavior data, personalized demand data, and complaint and churn information; the store information data module includes store product information, member information, order information, sales data, and customer behavior data. For the store information data, the products can be classified and summarized into primary, secondary and tertiary levels. The primary level includes gourmet food and beverages, department stores and daily necessities, the secondary level includes gourmet food and beverages such as snacks, vegetables, fruits, grains and oils, and the tertiary level includes grains and oils such as rice, noodles, oil, etc.

[0016] Preferably, the store information data module establishes and improves the customer database, so that the enterprise can achieve refined management of customer information, thereby improving customer satisfaction and loyalty, and analyzing key indicator data such as the number of visitors and click-through rate. The enterprise can adjust product prices and shipping templates to meet the needs of different customer groups. According to the customer's purchase history and preferences, the enterprise can recommend related products or services to improve conversion rate and customer satisfaction.

[0017] Preferably, the construction of the database module includes the following steps:

[0018] S301: Under the premise of legality and security of the requirements of GDPR, clarify the database requirements, including data type, data volume and performance requirements;

[0019] S302: Determine the basic structure of the data model, including entities, relationships, and attributes, and further refine the data model, including defining data tables, fields, and their attributes;

[0020] S303: Select a suitable database management system, determine the storage location of data files, design index and partition strategies, use DBMS to create an actual database, and import data into the database;

[0021] S304: After the database is built, comprehensive testing is required, including unit testing, integration testing, and performance testing, to ensure the correctness and stability of the database. The database also needs to be maintained and updated regularly to adapt to changes in business needs.

[0022] Preferably, for the database module, in order to improve database performance and accuracy, training the database includes the following steps:

[0023] S311: By analyzing historical query answers to obtain a knowledge database about future queries, combined with self-supervised learning methods and new loss functions for training, the response speed and accuracy of the database can be significantly improved;

[0024] S312: Apply the query optimizer to the optimization of the database. At the same time, by collecting database parameter data and system IO performance indicator data and using the SOFTMAX classifier for model training, the future database IO performance can be predicted.

[0025] S313: Introducing statistical constraints can detect potential errors that are difficult to detect with conventional constraints, thereby improving the correctness of the database. When building an analysis pipeline, automatically selecting the best-fit training data with high similarity to the input data can improve the quality of analysis.

[0026] Preferably, the storage module includes a preprocessing module, a preservation module and a protection module. The preprocessing module is used for data cleaning and preprocessing before data storage, such as removing duplicate data, solving data inconsistency and other problems to ensure data quality; for the preservation module, disk storage can be used, cloud computing services such as AWS, Azure and Google Cloud can be used for elastic storage and backup, object storage OSS can be used for large-scale data storage, and CDN can be combined to accelerate the distribution of static resources. The classification and summary information of the goods can also be stored separately, which is convenient for subsequent users to browse and purchase.

[0027] Preferably, the storage module classifies the data according to its importance and access frequency, stores frequently accessed data on a high-performance storage medium, and stores less frequently accessed data on a low-cost storage medium;

[0028] For the protection module, encryption technology and key management are used to protect the security and integrity of data. Data is backed up regularly to ensure that there are at least three copies of data and store them on two different types of media to prevent data loss. Data can also be backed up and archived to an offline location to avoid the risk of connecting to an attacked system or network.

[0029] Preferably, a deep learning-based dialogue system can be introduced for the communication module, which can understand and answer customers' questions through natural language processing technology, and continuously optimize its response strategy through historical dialogue analysis, using modern programming technology and database technology, including basic customer service functions, after-sales service order management, spare parts application and delivery functions.

[0030] According to another aspect of the present invention, there is provided a customer relationship management method, comprising the following steps:

[0031] S01: Collect personal information through the user information module, compare and classify it with the customer resource data of the database module, and store it through the storage module;

[0032] S02: Based on the results of data analysis, customers are segmented into different groups and personalized products and services are provided to each group. Customer segmentation helps companies carry out marketing activities more targeted and improve marketing effectiveness.

[0033] S03: accurately push the third-level data of the frequently purchased and browsed items according to the purchase records and preferences filled in by the customer. If the third-level data does not contain the item the user wants to buy, push the item data of the same category in the second-level data. If the second-level data does not contain the item the user wants to buy, push the item data of the same category in the first-level data. When the first-level to third-level data do not match, feedback the information to the database module (3) to re-acquire and analyze the data.

[0034] S04: Communicate with customers through the communication module, respond to user needs and problems in a timely manner, improve customer satisfaction and loyalty, and encourage customers to make repeat purchases and recommend new customers through reward mechanisms such as point systems, preferential activities, and membership systems, thereby improving customer loyalty. By communicating with customers and their browsing records, timely update customers' preferences and purchase needs, so that accurate push notifications can be made when customers browse and purchase next time;

[0035] S05: The communication module can also timely understand customer needs, pay attention to customer feedback, solve customer problems in a timely manner, and optimize products and services based on feedback, improve customer satisfaction, establish a customer churn early warning system, monitor customer behavior in real time, and promptly identify potential churn risks.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention can conveniently understand user information through the user information module, and cooperate with the database module to facilitate the classification of users, facilitate the push of suitable products, and facilitate the purchase and understanding of product information. By effectively managing customer information and needs, it can improve customer satisfaction and loyalty, which not only helps to retain old customers, but also attract new customers. In the fierce market competition, companies with strong customer relationship management capabilities can better meet customer needs and provide better services, thereby gaining competitive advantages in the market. By optimizing business processes and improving resource utilization efficiency, it can help companies reduce operating costs and improve work efficiency;

[0038] 2. The storage module can be used to store various information, clean and pre-process data to ensure data quality, protect and back up data to ensure security and integrity to prevent data loss, and back up and archive data to offline locations to avoid the risk of connecting to attacked systems or networks;

[0039] 3. The communication module is a multifaceted and multi-level process, involving data management, service quality improvement, technical support, user experience and other aspects. Through continuous technological innovation and strategy adjustment, the efficiency and quality of e-commerce customer service can be effectively improved, thereby enhancing customer satisfaction and corporate competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of the structure of a customer relationship management system proposed in Embodiment 1 of the present invention;

[0041] Figure 2 This is a schematic diagram of the structure of a database module in a customer relationship management system proposed in the first embodiment of the present invention;

[0042] Figure 3 This is a structural diagram of a storage module in a customer relationship management system proposed in Embodiment 1 of the present invention.

[0043] In the figure: 1. Server module; 2. User information module; 3. Database module; 31. Customer resource data module; 32. Store information data module; 4. Storage module; 41. Preprocessing module; 42. Preservation module; 43. Protection module; 5. Communication module. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Embodiment 1

[0046] See also Figure 1-Figure 3 , an embodiment provided by the present invention: a customer relationship management system, including a server module 1, a user information module 2, a database module 3, a storage module 4 and a communication module 5, the server module 1 is used for the architecture service between the modules, which is convenient for the management of customer relationships, and can build a platform for the server to facilitate communication and transactions between users and stores, etc., which can be a computer terminal or a mobile phone terminal, etc., which is a relatively mature technology and will not be described in detail here; the user information module 2 is used to collect and screen the basic information of the customer, which can facilitate the user to register a platform account, facilitate the push of suitable products according to needs, and facilitate the purchase and understanding of product information; the database module 3 is used to establish an information database for the information of customer resource data and store information data, which can facilitate the classification of users, and facilitate the push of products that meet their own needs during subsequent purchases, etc., which is more convenient and quick; the storage module 4 is used to store user information, transaction records, communication information, and store information; the communication module 5 is used for communication in preliminary consultation, sales communication, and after-sales service.

[0047] Furthermore, the user information module 2 includes personal identity information and network activity information;

[0048] Personal identity information mainly includes the user's name, date of birth, ID number, address, telephone number, account number and password, and preferences. It can facilitate real-name authentication to ensure the security of subsequent transactions. It can also facilitate the preferences and needs filled in in the early stage to facilitate the push of suitable stores in the later stage, making it easier to purchase and understand suitable products. Network activity information includes the user's IP address, Cookie information, online and technical information such as Internet or other electronic network activity information, and the user's fingerprints, iris and other information can also be entered to facilitate the authentication of subsequent transactions and ensure security.

[0049] Furthermore, the database module 3 includes a customer resource data module 31 and a store information data module 32. The customer resource module 31 includes social characteristic data, such as name, address, telephone number, Internet website, etc., purchase behavior data including the type, quantity, amount and frequency of purchased services, personalized demand data such as special requirements and habits, and complaint and churn information, recording customer complaints and whether customers have churned. The customer resource data can be mainly obtained through customer registration, registration, survey, and service operation records, including each interaction record with the customer, customer feedback and complaint information on the service. The social characteristic data mainly collects the user information of the user information module 2 to establish a database, which is convenient for subsequent purchase and browsing, and then pushes suitable products.

[0050] The store information data module 32 includes store product information, member information, order information, sales data, and customer behavior data. The store information database may include multiple data tables, such as a member information table and an order information table, for convenient record keeping and statistics. By establishing and improving a customer database, the store information data module 32 enables enterprises to achieve refined management of customer information, thereby improving customer satisfaction and loyalty, and analyzing key indicator data such as the number of visitors and click-through rate. Enterprises can adjust product prices and shipping fee templates to meet the needs of different customer groups. Based on the customer's purchase history and preferences, enterprises can recommend related products or services to improve conversion rates and customer satisfaction.

[0051] For the store information data, the goods can be classified and summarized into the first level, the second level and the third level. The first level includes food and beverages, daily necessities, etc. The second level includes snacks, vegetables, fruits, grains and oils for food and beverages, and the third level includes rice, noodles, oil, etc. for grains and oils. Among them, the first level is not limited to food and beverages and daily necessities, but also includes clothing, tools, home appliances, etc. The corresponding first-level classification also has corresponding second and third levels, which is convenient for classification and induction. The induction is set according to the situation of the store. For rice, noodles, and oil in grains and oils, it can also include cereals, etc., and there can also be fourth or fifth levels, etc. For example, for oil, it can be soybean oil, olive oil, corn oil, etc., and the fifth level for soybean oil can be pressed oil, leached oil, genetically modified and non-genetically modified, etc. Similar classification can be done for other commodities.

[0052] Furthermore, the construction of the database module 3 includes the following steps:

[0053] S301: Under the premise of legality and security of the requirements of GDPR, clarify the database requirements, including data type, data volume and performance requirements;

[0054] S302: Determine the basic structure of the data model, including entities, relationships, and attributes, provide a high-level view to help understand the basic connections between data, and further refine the data model, including defining data tables, fields, and their attributes to avoid data redundancy and update anomalies.

[0055] S303: Select a suitable database management system, such as DBMS, determine the storage location of data files, design index and partition strategies, use DBMS to create an actual database, import data into the database, optimize database performance and storage efficiency, ensure database security, back up data, set recovery strategies, etc.;

[0056] S304: After the database is built, comprehensive testing is required, including unit testing, integration testing, and performance testing, to ensure the correctness and stability of the database. The database also needs to be maintained and updated regularly to adapt to changes in business needs.

[0057] For database construction, ensuring that the collected and managed data is accurate is the basis for improving the quality of research results. During the design and development phase, the logical and physical structure of the database needs to be optimized to meet specific business needs and achieve optimal system performance, comply with relevant laws and regulations, and take effective measures to protect the security and privacy of personal data. The scalability and flexibility of the database should be considered during design to facilitate future adjustments and upgrades based on changes in business needs.

[0058] Furthermore, for the database module 3, in order to improve the database performance and accuracy, training the database includes the following steps:

[0059] S311: By analyzing historical query answers to obtain a knowledge database about future queries, the answer can be generated by maximizing the entropy principle. Combining self-supervised learning methods such as GNN with new loss functions such as InfoNode can significantly improve the response speed and accuracy of the database;

[0060] S312: Apply a query optimizer, such as QO, to optimize the database. At the same time, by collecting database parameter data and system IO performance indicator data and using a SOFTMAX classifier for model training, future database IO performance can be predicted.

[0061] S313: Introducing statistical constraints can detect potential errors that are difficult to detect with conventional constraints, thereby improving the correctness of the database. When building an analysis pipeline, automatically selecting the best-fit training data with high similarity to the input data can improve the quality of analysis.

[0062] Improving database performance and accuracy requires comprehensive consideration of data accuracy, the use of advanced learning mechanisms, optimization using machine learning techniques, and the selection of appropriate training data and optimization strategies. Through these methods, the performance and accuracy of the database can be effectively improved.

[0063] Furthermore, the storage module 4 includes a pre-processing module 41, a preservation module 42 and a protection module 43. The pre-processing module 41 is used to clean and pre-process data before storage, such as removing duplicate data, solving data inconsistency and other problems to ensure data quality, and using RAID technology, space guarantee function and flexible volume configuration to optimize storage efficiency and manageability; the preservation module 42 can use disk storage, use cloud computing services such as AWS, Azure and Google Cloud for elastic storage and backup, use object storage OSS for large-scale data storage, and combine CDN to accelerate the distribution of static resources;

[0064] The storage module 4 can also store the classification and summary information of the goods separately, so as to facilitate the subsequent users to browse and purchase, and to read and push. At the same time, it can also store the user's preferences, browsing and purchase information, so as to facilitate the subsequent purchase and browsing, and to facilitate accurate push and communication. It can also cooperate with the database module 3 to store the classification and summary information of the goods and the correlation degree of customer resource data in different files, so as to facilitate the subsequent users to browse and purchase, and to read and push.

[0065] The following steps are included for storing the association level:

[0066] S31: Extract all relevant features or indicators from the data, and then perform dimensionless processing on the data to eliminate different dimensions and orders of magnitude of different features;

[0067] S32: Use grey correlation analysis to calculate the correlation between each feature, establish a correlation matrix, and evaluate the similarity between each feature by calculating the correlation coefficient;

[0068] S33: According to the size of the correlation, the features are classified into several levels, such as high, medium, and low, and the features are assigned to corresponding levels according to the size of their correlation. When the user browses or purchases the products in the future, the products are pushed according to the correlation.

[0069] For the storage module 42, data is classified according to its importance and access frequency, and frequently accessed data is stored on high-performance storage media, while less frequently accessed data is stored on low-cost storage media. By capturing the time point image of the data volume, additional storage space is consumed only when the volume changes, thereby achieving efficient storage space utilization. For big data applications, distributed database systems such as Hadoop HDFS and NoSQL databases can be used to effectively manage and retrieve data;

[0070] For the protection module 43, encryption technology and key management are used to protect the security and integrity of data. Data is backed up regularly to ensure that there are at least three copies of data and store them on two different types of media to prevent data loss. Data can also be backed up and archived to an offline location to avoid the risk of connecting to an attacked system or network.

[0071] Furthermore, for the communication module 5, a deep learning-based dialogue system can be introduced, which can understand and answer customers' questions through natural language processing technology, and continuously optimize its response strategy through historical dialogue analysis, using modern programming technology and database technology, such as .NET platform and SQL Server, including basic customer service functions, after-sales service order management, spare parts application and delivery functions. Intelligent assistants can also be introduced. For example, the AliMe intelligent assistant provides services in the form of question-and-answer pairs, which can handle a large number of customer issues. At the same time, knowledge reasoning is performed through event graphs to improve the accuracy and efficiency of problem solving.

[0072] Embodiment 2

[0073] A customer relationship management method comprises the following steps:

[0074] S01: Register and fill in personal information through the user information module 2, compare and classify it with the customer resource data of the database module 3, and store it through the storage module 4;

[0075] S02: Based on the results of data analysis, customers are segmented into different groups and personalized products and services are provided to each group. Customer segmentation helps companies carry out marketing activities more targeted and improve marketing effectiveness.

[0076] S03: Accurately push the third-level data of frequently purchased and browsed products based on the customer's purchase records and preferences, so as to facilitate accurate browsing and purchase by customers and improve purchase efficiency. If there is no product that the user needs to buy in the third-level data, the product data of the same category in the second-level data will be pushed. When pushing products of the corresponding level, the proportion of main purchases is 70-80%, and the proportion of similar products of the corresponding level is 20-30%, which can facilitate users' purchase selection;

[0077] S04: Communicate with customers through communication module 5, respond to user needs and problems in a timely manner, improve customer satisfaction and loyalty, encourage customers to make repeated purchases and recommend new customers through reward mechanisms such as point systems, preferential activities, and membership systems, thereby improving customer loyalty. By communicating with customers and their browsing records, timely update customers' preferences and purchase needs, so that they can be accurately pushed when they browse and buy next time. For customer data updates, the purchased and browsed products can be updated and saved every 7 days, and surveys can be conducted every one to two months. Questionnaires and other methods can be used to understand users' purchasing needs in a timely manner, so that they can be accurately and quickly pushed when they make subsequent purchases, thereby improving users' comfort when using the product.

[0078] S05: Communication module 5 can also timely understand customer needs, pay attention to customer feedback, solve customer problems in a timely manner, and optimize products and services based on feedback, improve customer satisfaction, establish a customer churn early warning system, monitor customer behavior in real time, and promptly identify potential churn risks.

[0079] In summary, the present invention can conveniently understand user information through the user information module 2 provided, and can conveniently classify users with the database module 3, conveniently push suitable products, and also conveniently purchase and understand product information. By effectively managing customer information and needs, customer satisfaction and loyalty can be improved, which not only helps to retain old customers, but also attract new customers. In the fierce market competition, companies with strong customer relationship management capabilities can better meet customer needs and provide better services, thereby gaining competitive advantages in the market. By optimizing business processes and improving resource utilization efficiency, it can help companies reduce operating costs and improve work efficiency; various information can be conveniently stored through the storage module 4, and data can be cleaned and preprocessed to ensure data quality, and data can be protected and backed up to ensure security and integrity to prevent data loss. Data can also be backed up and archived to an offline location to avoid the risk of connecting to an attacked system or network; the communication module 5 is a multi-faceted and multi-level process involving data management, service quality improvement, technical support, and user experience. Through continuous technological innovation and strategy adjustment, the efficiency and quality of e-commerce customer service can be effectively improved, thereby enhancing customer satisfaction and corporate competitiveness.

[0080] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A customer relationship management system, characterized in that: It comprises a server module (1), a user information module (2), a database module (3), a storage module (4) and a communication module (5), wherein the server module (1) is used for providing architecture services between the modules to facilitate the management of customer relationships; The user information module (2) is used to collect and screen the basic information of the customer; The database module (3) is used to establish an information base based on customer resource data and store information data, wherein the store information data includes a classification of commodities, and the commodities are classified into primary, secondary and tertiary categories; The storage module (4) is used to store user information, transaction records, communication information and store information data, and cooperates with the database module (3) to store the classification and summary information of commodities and the correlation degree of customer resource data in different files, so as to facilitate subsequent users to browse and push when purchasing; The storage module (4) comprises a pre-processing module (41), a storage module (42) and a protection module (43). The preprocessing module (41) is used to clean and preprocess the data before storage to remove duplicate data; The storage module (42) uses disk storage and cloud computing services for elastic storage and backup, uses object storage OSS for large-scale data storage, and combines CDN to accelerate the distribution of static resources; The storage module (42) classifies the data according to its importance and access frequency, stores the frequently accessed data in a high-performance storage medium, and stores the less frequently accessed data in a low-cost storage medium; The protection module (43) uses encryption technology and key management to protect the security and integrity of data, backs up data regularly, ensures that there are at least three copies of data, and stores them on two different types of media to prevent data loss, and backs up and archives data to an offline location to avoid the risk of connecting to a compromised system or network; The customer resource data relevance classification and storage includes the following steps: S31: extract all relevant features or indicators from customer resource data, perform dimensionless processing, and eliminate different dimensions and orders of magnitude of different features; S32: Use grey correlation analysis to calculate the correlation between each feature, establish a correlation matrix, and evaluate the similarity between each feature by calculating the correlation coefficient; S33: Classify the features according to the relevance, set several levels, and assign the features to corresponding levels according to their relevance. When the user browses or purchases the products, the products are pushed according to the relevance. The communication module (5) is used for early consultation, sales communication and after-sales service communication.

2. A customer relationship management system according to claim 1, characterized in that: The user information module (2) includes personal identity information and network activity information; The personally identifiable information includes the user's name, date of birth, address, phone number, account number and password, and preferences; The network activity information includes the user's IP address, cookie information, online and technical information.

3. A customer relationship management system according to claim 2, characterized in that: The database module (3) includes a customer resource data module (31) and a store information data module (32). The customer resource module (31) includes social characteristic data, purchase behavior data, personalized demand data, complaints and churn information. The social characteristic data includes a database established based on the information collected by the user information module (2), which is convenient for subsequent purchase and browsing, and then pushes suitable products; The store information data module (32) includes store product information, member information, order information, sales data and customer behavior data.

4. A customer relationship management system according to claim 3, characterized in that: The store information data module (32) establishes and improves a customer database, analyzes the database data, adjusts product prices and shipping fee templates to meet the needs of different customer groups, analyzes customer purchase history and preferences, recommends related products or services, and improves conversion rates and customer satisfaction.

5. A customer relationship management system according to claim 4, characterized in that: The construction of the database module (3) includes the following steps: S301: Under the premise of legality and security of the requirements of GDPR, the database requirements should be clarified, including data type, data volume and performance requirements; S302: Determine the basic structure of the data model, including entities, relationships, and attributes, and further refine the data model, including defining data tables, fields, and their attributes; S303: Select a suitable database management system, determine the storage location of data files, design index and partition strategies, use DBMS to create an actual database, and import data into the database; S304: After the database is built, comprehensive testing is carried out, including unit testing, integration testing and performance testing. The database is regularly maintained and updated to adapt to changes in business needs.

6. A customer relationship management system according to claim 5, characterized in that: The database module (3) trains the database including the following steps: S311: Obtain a knowledge database about future queries by analyzing historical query answers, and train with a self-supervised learning method and loss function; S312: Optimize the database using a query optimizer, collect database parameter data and system IO performance indicator data, and use a SOFTMAX classifier to train a model to predict future database IO performance; S313: Introduce statistical constraints to detect potential errors that are difficult to detect with conventional constraints. When building an analysis pipeline, automatically select the best-fit training data with high similarity to the input data for training.

7. A customer relationship management system according to claim 6, characterized in that: The communication module (5) introduces a deep learning-based dialogue system, which uses natural language processing technology to understand and answer customers' questions. It also continuously optimizes its response strategy through historical dialogue analysis and uses modern programming and database technologies to implement basic customer service functions, after-sales service order management, spare parts application and delivery functions.

8. A customer relationship management method, used for use in a customer relationship management system according to claim 7, characterized in that: The following steps are involved: S01: collecting personal information through the user information module (2), comparing and classifying it with the customer resource data of the database module (3), and storing it through the storage module (4); S02: Based on the results of data analysis, customers are segmented into different groups and personalized products and services are provided to each group; S03: accurately push the third-level data of the frequently purchased and browsed items according to the purchase records and preferences filled in by the customer. If the third-level data does not contain the item the user wants to buy, push the item data of the same category in the second-level data. If the second-level data does not contain the item the user wants to buy, push the item data of the same category in the first-level data. When the first-level to third-level data do not match, feedback the information to the database module (3) to re-acquire and analyze the data. S04: Communicate with customers through the communication module (5), respond to user needs and questions in a timely manner, and update customer preferences and purchase needs in a timely manner through customer communication records and customer browsing records, so that accurate push notifications can be made the next time the customer browses and purchases.

Images