Method, device and equipment for interoperation of heterogeneous data of retail enterprises and storage medium

Abstract

The application discloses a retail enterprise heterogeneous data interoperability method and device, equipment and storage medium, relates to the technical field of data processing, and comprises the following steps: connecting each heterogeneous database of a target retail enterprise by using a preset connector, reading operation data in each heterogeneous database to obtain each target heterogeneous data, performing semantic analysis and data correlation analysis on each target heterogeneous data by using a preset artificial intelligence technology to determine a target entity corresponding to each target heterogeneous data, and constructing a preset logical data model based on the target entity; obtaining an information interoperability request sent by a data interoperability request party to the target retail enterprise, and converting the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so that the operation data in each heterogeneous database is subjected to data interoperability based on the target interoperability instructions to obtain heterogeneous data interoperability results of the target retail enterprise. In this way, the data interoperability can be improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to interoperability methods, apparatus, equipment and storage media for heterogeneous data in retail enterprises. Background Technology

[0002] With the deepening of digital transformation, enterprise data is experiencing explosive growth, and its sources are becoming increasingly diversified. Traditional relational databases, various NoSQL databases (Not Only SQL, i.e., non-relational databases), real-time streaming data, and a large number of semi-structured and unstructured files coexist, forming an extremely complex heterogeneous data environment. Against this backdrop, different business systems and departments often independently build their data storage and management systems according to their own needs, resulting in significant differences in data format, structure, and semantics, leading to severe "data fragmentation" and "data silos." This dispersed and heterogeneous state makes it difficult for enterprises to build a global, unified data view, hindering not only efficient cross-departmental and cross-business collaboration but also directly impacting the development of in-depth data insights and intelligent decision-making capabilities.

[0003] Currently, mainstream data integration solutions still heavily rely on manual sorting, manual configuration of mapping rules, and hard-coded data transformation logic. When faced with massive, continuously growing, and dynamically changing heterogeneous data sources, these traditional methods reveal significant limitations: long implementation cycles, high manpower investment, and exorbitant maintenance costs. More importantly, their inherent static and rigid nature makes them ill-suited to the dynamic demands of rapidly evolving business needs, such as changes in data patterns and the integration of new data sources. Once the source data structure or business rules change, cumbersome manual intervention and code modifications are often required, lacking sufficient agility and adaptability.

[0004] Therefore, it is necessary to solve the problem of improving the response speed and analysis efficiency of cross-source queries when facing massive heterogeneous data sources of retail enterprises. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide an interoperability method, apparatus, device, and storage medium for heterogeneous data in retail enterprises, which can improve the response speed and analysis efficiency of cross-source queries. The specific solution is as follows: In a first aspect, this application discloses an interoperability method for heterogeneous data in retail enterprises, applied to a pre-defined heterogeneous data interoperability engine, comprising: The system connects to the heterogeneous databases of the target retail enterprise using a pre-defined connector and reads the operational data from each of the heterogeneous databases to obtain the target heterogeneous data; the heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise. Using preset artificial intelligence technology, semantic analysis and data association analysis are performed on each of the target heterogeneous data to determine the target entity corresponding to each of the target heterogeneous data, and a preset logical data model is constructed based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data. The system acquires the information interoperability request sent by the data interoperability requester to the target retail enterprise, and transforms the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

[0006] Optionally, the step of connecting to the heterogeneous databases of the target retail enterprise using a preset connector and reading operational data from each of the heterogeneous databases to obtain target heterogeneous data includes: For each heterogeneous database of the target retail enterprise, a preset connector is determined for each heterogeneous database; the preset connector establishes a corresponding connection with each heterogeneous database of the target retail enterprise. The operation data in each of the heterogeneous databases is read using the preset connector and based on the preset data reading method to obtain the target heterogeneous data; the preset data reading method includes full synchronization, incremental synchronization and real-time streaming data ingestion.

[0007] Optionally, the step of connecting to the heterogeneous databases of the target retail enterprise using a preset connector and reading operational data from each of the heterogeneous databases to obtain target heterogeneous data includes: The system connects to the heterogeneous databases of the target retail enterprise using a pre-defined connector, and reads the operational data from each of the heterogeneous databases using a pre-defined data transmission encryption method to obtain the initial heterogeneous data. The initial heterogeneous data are subjected to data format conversion, character set encoding unification, data type conversion and data cleaning to obtain the target heterogeneous data.

[0008] Optionally, the step of using preset artificial intelligence technology to perform semantic analysis and data association analysis on each of the target heterogeneous data to determine the target entity corresponding to each of the target heterogeneous data, and constructing a preset logical data model based on the target entity, includes: Natural language processing techniques are used to analyze the semantic relationships of text metadata in each of the target heterogeneous data to obtain the first analysis result; Clustering algorithms are used to analyze the data patterns and data in each of the target heterogeneous data sets to obtain the dataset association relationship between the target heterogeneous data sets, and the dataset association relationship is determined as the second analysis result; Utilize knowledge graph construction techniques to build a target semantic model; Based on the first analysis result and the second analysis result, the target entities corresponding to each of the target heterogeneous data and the data mapping relationship between each of the target heterogeneous data are determined; A preset logical data model is constructed based on the target entity, the data mapping relationship, and the target semantic model.

[0009] Optionally, the process of acquiring the data interoperability requester's information interoperability request sent to the target retail enterprise, and converting the information interoperability request into target interoperability instructions corresponding to each of the heterogeneous databases based on the logical data model, includes: Based on the preset query interface, the data interoperability requester obtains the information interoperability request sent by the data interoperability requester to the target retail enterprise. The logical data model is used to parse the information interoperability request in order to determine the target database corresponding to the information interoperability request from each of the heterogeneous databases; Based on the data mapping relationships stored in the logical data model, the information interoperability request is transformed into the target interoperability instruction corresponding to the target database.

[0010] Optionally, after performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instructions, the method further includes: The target data corresponding to the data interoperation is analyzed to determine whether the target data contains sensitive data; the sensitive data is data that the data interoperation requester is prohibited from accessing in advance. If the target data contains the sensitive data, then the sensitive data is desensitized using a data desensitization method to obtain the desensitized target data; the data desensitization method includes data masking, data generalization, data truncation, data perturbation, and cryptographic hashing.

[0011] Optionally, after constructing the preset logical data model based on the target entity, the method further includes: The data mapping relationship stored in the preset logical data model is displayed on the visualization canvas of the preset visualization interface, so that the user can modify and adjust the data mapping relationship on the visualization canvas; Accordingly, after performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instructions, the method further includes: The data access results corresponding to the data interoperation are processed into images to generate corresponding data visualization tables, and the data visualization tables are displayed on the preset visualization interface.

[0012] Secondly, this application discloses an interoperability device for heterogeneous data in retail enterprises, applied to a pre-defined heterogeneous data interoperability engine, comprising: The data reading module is used to connect with the heterogeneous databases of the target retail enterprise using a preset connector, and read the operational data in each of the heterogeneous databases to obtain the target heterogeneous data; the heterogeneous databases are deployed in the corresponding operational scenarios of the target retail enterprise. The data analysis module is used to perform semantic analysis and data association analysis on each of the target heterogeneous data using preset artificial intelligence technology, so as to determine the target entity corresponding to each of the target heterogeneous data, and construct a preset logical data model based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data. The data interoperability module is used to acquire information interoperability requests sent by the data interoperability requester to the target retail enterprise, and to convert the information interoperability requests into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

[0013] Thirdly, this application discloses an electronic device, comprising: Memory, used to store computer programs; A processor for executing the computer program to implement the aforementioned interoperability method for heterogeneous data in retail enterprises.

[0014] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the aforementioned interoperability method for heterogeneous data of retail enterprises.

[0015] As can be seen, in this application, a preset connector is used to connect with various heterogeneous databases of the target retail enterprise, and operational data from each heterogeneous database is read to obtain target heterogeneous data. The heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise. Preset artificial intelligence technology is used to perform semantic analysis and data association analysis on each target heterogeneous data to determine the target entities corresponding to each target heterogeneous data, and a preset logical data model is constructed based on the target entities. The preset logical data model stores the data mapping relationship between the target entities and each target heterogeneous data. Information interoperability requests sent by data interoperability requesters to the target retail enterprise are obtained, and based on the logical data model, the information interoperability requests are converted into target interoperability instructions corresponding to each heterogeneous database, so that data interoperability is performed on the operational data in each heterogeneous database based on the target interoperability instructions to obtain the corresponding heterogeneous data interoperability results of the target retail enterprise. That is, this application uses a preset connector to connect with various heterogeneous databases corresponding to the target retail enterprise to read operational data from each heterogeneous database, then uses preset artificial intelligence technology to analyze the data association relationships between each operational data, and then constructs a preset logical data model. During data interoperability, the obtained information interoperability requests can be transformed into target interoperability instructions corresponding to each heterogeneous database through a pre-defined logical data model, and then the corresponding target interoperability instructions can be executed. This ensures seamless compatibility with diverse data sources during data interoperability. Based on this, through intelligent data pattern recognition and relationship mapping, and utilizing machine learning and artificial intelligence technologies, the bottlenecks of traditional manual integration methods are overcome, achieving automated understanding and fusion of data semantics and structure. This not only significantly reduces data integration costs but also improves the accuracy and adaptability of data integration. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This application discloses a flowchart of a method for interoperating heterogeneous data in a retail enterprise. Figure 2 This is a schematic diagram of the interoperability device for heterogeneous data in a retail enterprise disclosed in this application. Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] With the surge in enterprise data volume and diversification of sources, heterogeneous data from various sources such as traditional databases, NoSQL, and streaming data are stored independently, resulting in inconsistent formats and semantics and creating severe data silos. Existing integration solutions mostly rely on manual configuration and hard coding, which are inefficient, costly, and poorly adaptable when processing massive amounts of dynamic data, making it difficult to support a global data view and agile analysis and decision-making. Therefore, this application will specifically introduce an interoperability method for heterogeneous data in retail enterprises, which can solve the above problems.

[0020] See Figure 1 As shown in the figure, this application discloses an interoperability method for heterogeneous data in retail enterprises, applied to a preset heterogeneous data interoperability engine, including: Step S11: Connect to the heterogeneous databases of the target retail enterprise using a preset connector, and read the operational data from each heterogeneous database to obtain the target heterogeneous data; the heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise.

[0021] In this embodiment, the step of connecting to the heterogeneous databases of the target retail enterprise using preset connectors and reading operational data from each heterogeneous database to obtain target heterogeneous data includes: determining preset connectors for each heterogeneous database based on the database type corresponding to each heterogeneous database of the target retail enterprise; establishing corresponding connections between the preset connectors and each heterogeneous database of the target retail enterprise; and using the preset connectors and based on preset data reading methods to read operational data from each heterogeneous database to obtain target heterogeneous data. The preset data reading methods include full synchronization, incremental synchronization, and real-time streaming data ingestion.

[0022] First, it should be noted that the "heterogeneity" mentioned in this application refers to the type of data source, data format, data structure, and data access method. For example, the data may come from traditional relational databases, which are usually well-structured; it may also come from NoSQL databases, which may use different data models such as key-value pairs, documents, column families, or graphs; in addition, there is unstructured data (such as text files, images, audio and video, and various document data) and semi-structured data (various code languages) in file systems; as well as various service data provided through API (Application Programming Interface); and even real-time streaming data (such as real-time information streams from sensors, IoT devices, and social media).

[0023] During operation, multi-source heterogeneous data is collected through pre-defined connectors. These connectors support various data source types, including but not limited to relational databases, various NoSQL databases, distributed file systems (such as HDFS (Hadoop Distributed File System) and Ceph), cloud storage services, web service APIs, and real-time data streams (such as Kafka (Apache Kafka) and Pulsar). During the data acquisition phase, the pre-defined connectors provide various connectors and protocol adapters, such as JDBC (Java Database Connectivity), ODBC (Open Database Connectivity), RESTful API (Representational State Transfer), and specific SDKs (Software Development Kits), to ensure seamless communication with different data sources. The acquisition process supports full synchronization, incremental synchronization (based on timestamps, version numbers, or log parsing), and real-time streaming data ingestion to meet the timeliness requirements of different business scenarios. To address the heterogeneity of the data, this module performs preliminary adaptation processing before the data enters the engine. This includes data format conversion (such as parsing text from various code languages ​​into an internally unified data structure), character set encoding standardization, basic data type conversion, and preliminary data cleaning (such as removing null values ​​and format validation). This module possesses robust error handling and retry mechanisms to ensure the reliability and integrity of data transmission. Thus, during data acquisition, appropriate format conversion, character set standardization, data type validation, and basic data cleaning are performed based on the characteristics of different data sources. For example, CSV (Comma-Separated Values) files are parsed into structured records, JSON (JavaScript Object Notation) or XML (Extensible Markup Language) data is converted into an internally unified data object representation, or timestamps are standardized. To ensure acquisition efficiency, this application employs various strategies such as batch processing, incremental synchronization, or real-time stream processing to adapt to different data volume requirements and data source characteristics.

[0024] In this embodiment, the step of connecting to the heterogeneous databases of the target retail enterprise using a preset connector and reading operational data from each heterogeneous database to obtain target heterogeneous data includes: connecting to the heterogeneous databases of the target retail enterprise using a preset connector and reading operational data from each heterogeneous database using a preset data transmission encryption method to obtain initial heterogeneous data; and performing data format conversion, character set encoding unification, data type conversion, and data cleaning on each initial heterogeneous data to obtain target heterogeneous data. Considering the sensitivity of the data, this embodiment has security mechanisms such as encrypted data transmission, identity authentication, and authorization to ensure that data is not leaked or tampered with during transmission from the source to the engine. Its design goal is to provide a flexible, scalable, and high-performance data access layer, laying the foundation for subsequent data integration and analysis, and effectively breaking down data silos. That is, to ensure data transmission security, this embodiment supports data transmission encryption (such as TLS / SSL) and source-end identity authentication to prevent data from being illegally intercepted or tampered with during the collection process. Its design goal is to build a robust, efficient, and flexible data access layer, laying a solid foundation for subsequent intelligent processing. Among them, data transmission encryption methods include, but are not limited to, TLS (Transport Layer Security) and SSL (Secure Socket Layer).

[0025] Step S12: Use preset artificial intelligence technology to perform semantic analysis and data association analysis on each of the target heterogeneous data to determine the target entity corresponding to each of the target heterogeneous data, and construct a preset logical data model based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data.

[0026] In big data environments, data from different data sources is often independently designed. They may use different naming conventions, data types, representations, and even have different interpretations of the same concept. For example, one database might represent a customer's name as "CustomerName," while another database might use "Client_FN" and "Client_LN." Traditional data integration methods typically rely on manually defined rules, which are inefficient and error-prone when dealing with large-scale heterogeneous data.

[0027] Therefore, in this embodiment, the step of using preset artificial intelligence technology to perform semantic analysis and data association analysis on each of the target heterogeneous data to determine the target entities corresponding to each of the target heterogeneous data, and constructing a preset logical data model based on the target entities, includes: using natural language processing technology to analyze the semantic association of text metadata in each of the target heterogeneous data to obtain a first analysis result; using clustering algorithms to analyze data patterns and data in each of the target heterogeneous data to obtain the dataset association relationship between the target heterogeneous data, and determining the dataset association relationship as a second analysis result; using knowledge graph construction technology to construct a target semantic model; determining the target entities corresponding to each of the target heterogeneous data and the data mapping relationship between each of the target heterogeneous data based on the first analysis result and the second analysis result; and constructing a preset logical data model based on the target entities, the data mapping relationship, and the target semantic model. That is, using natural language processing technology to analyze the semantic association of text metadata such as field names and table names; using clustering algorithms to identify columns with similar patterns or content; using association rule mining algorithms to discover potential associations between different datasets; and using knowledge graph construction technology to construct a unified semantic model. Specifically, the system automatically identifies entities (such as "customers" and "orders") from different data sources using the aforementioned technologies and understands the relationships between these entities. It then intelligently infers data types, identifies primary and foreign key relationships, and discovers potential data dependencies and constraints. Based on data relationship identification, a globally unified logical data model (or conceptual model / unified meta-model) is constructed. This model abstracts the physical storage details of the underlying heterogeneous data sources, providing a semantically consistent view. Subsequently, mapping rules from heterogeneous data sources to this logical model are automatically generated. This includes field renaming, data type conversion, multi-field merging, single-field splitting, and the generation of complex conversion functions. For uncertain or low-confidence mappings, the system provides an interactive interface, allowing users to confirm, correct, or customize rules. That is, the data mapping relationships stored in the preset logical data model are displayed on a visual canvas on a preset visualization interface, allowing users to modify and adjust these relationships on the visualization canvas. In this way, through continuous learning and feedback mechanisms, the accuracy of pattern recognition and mapping can be continuously optimized, thereby achieving efficient and low-cost cross-source data integration and significantly improving data semantic interoperability. In other words, by introducing machine learning and artificial intelligence technologies, pattern recognition is automated and intelligent. It automatically analyzes metadata (such as table structure, field names, data types, constraints, etc.) and actual data content (such as data distribution, value range, and relationships) from different data sources to discover potential data patterns, entity relationships, and semantic information. For example, it identifies similar columns through clustering algorithms, understands the meaning of field names through natural language processing, and discovers implicit connections between data through association rule mining.Based on this, a unified logical data model can be intelligently constructed. This logical model is an abstract, high-level data view that integrates information from all heterogeneous data sources and eliminates the differences between them.

[0028] In the era of big data, traditional single-server storage architectures can no longer meet the requirements of data volume, concurrent access, and fault tolerance. Therefore, adopting distributed storage technology is an inevitable choice. HDFS, with its high throughput, high fault tolerance, and large file storage capabilities, is very suitable for storing massive amounts of unstructured and semi-structured data. It provides fault tolerance by distributing data blocks across multiple nodes in the cluster and replicating data blocks, ensuring data availability even if some nodes fail. NoSQL databases such as HBase provide high availability, linear scalability, and low-latency random read / write capabilities, suitable for storing large amounts of key-value pairs, column families, or document-type data, especially suitable for scenarios requiring fast lookups and updates. Therefore, this application will intelligently select or combine different storage technologies based on the characteristics of the data (structured, semi-structured, unstructured, access patterns, etc.) to achieve optimal storage efficiency and performance. In addition to simple storage, management functions are equally important. This includes data version control, lifecycle management (e.g., automatic archiving or deletion of old data), data backup and recovery, and metadata management. Metadata management refers to the management of the "data" itself, such as information like data source, format, owner, creation time, update time, and security level. This metadata is crucial for data retrieval, understanding, and governance. Distributed storage can easily handle explosive growth in data volume, supporting rapid data read, write, and retrieval, providing a solid foundation for upper-level data analysis and querying. It ensures high availability and durability of data; even in the event of partial hardware failure, data remains intact and accessible, serving as the cornerstone of the entire engine's stable operation. It fully leverages the advantages of big data storage technologies, such as the Hadoop Distributed File System (HDFS, HBase), to build the underlying data lake or data warehouse. Specifically, mature big data storage technologies, such as the Hadoop Distributed File System, are used to store large amounts of unstructured and semi-structured data, providing high throughput and high fault tolerance; distributed NoSQL databases like HBase are used to store structured or semi-structured key-value pairs, documents, or wide-column data to support low-latency random read / write and high availability; simultaneously, distributed relational databases or data warehouses can be integrated for efficient querying and analysis of complex structured data. This allows data to be allocated to the most suitable storage media and systems based on its access frequency, importance, and structural characteristics. For example, hot data can be stored in high-performance NoSQL databases, warm data in HDFS, and cold data can be archived to lower-cost cloud storage. Management functions cover the entire data lifecycle, including storage strategy selection after data ingestion, data replica management to ensure high availability and disaster recovery capabilities, data version control to support historical data backtracking, and data archiving and deletion strategies.Furthermore, in this embodiment, the source, format, owner, creation / modification time, security level, and relationship with other data of each data object are meticulously recorded and indexed, providing rich data context information for upper-layer modules, which greatly improves the manageability, discoverability, and reliability of the data, and ensures that the engine can efficiently process large amounts of data.

[0029] Step S13: Obtain the information interoperability request sent by the data interoperability requester to the target retail enterprise, and convert the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

[0030] In this embodiment, the process of acquiring the information interoperability request sent by the data interoperability requester to the target retail enterprise and converting the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model includes: acquiring the information interoperability request sent by the data interoperability requester to the target retail enterprise based on a preset query interface; parsing the information interoperability request using the logical data model to determine the target database corresponding to the information interoperability request from each heterogeneous database; and converting the information interoperability request into target interoperability instructions corresponding to the target database based on the data mapping relationship stored in the logical data model. Specifically, a unified query interface is first provided, such as a query language based on standard SQL or SQL-like languages, so that the data interoperability requester does not need to care about the specific implementation details of the underlying data source. When the data interoperability requester submits a query, this application will perform query parsing and convert it into an internal logical query plan. The next key step is the automatic translation function. Since the underlying heterogeneous data sources may use different query languages ​​(such as SQL, NoSQL query languages, graph query languages, etc.), it is necessary to intelligently decompose the unified logical query plan and convert it into query syntax specific to each data source. This requires a deep understanding of the query capabilities and syntax characteristics of various data sources. More importantly, query optimization is key to improving query performance. These optimization algorithms consider multiple factors, including data distribution, indexing information, network latency, data source processing capabilities, and data transfer costs. For example, it might intelligently select the optimal query execution order, decide which data source to perform filtering operations on to reduce the amount of data transferred, or utilize parallel processing techniques to execute subqueries simultaneously on multiple data sources. Furthermore, it can employ techniques such as query caching, materialized views, and pre-computation to accelerate common queries.

[0031] Specifically, a unified query interface is provided externally, such as an interface supporting the ANSI SQL standard, so that data interoperability requesters do not need to understand the specific query language and data structure of the underlying heterogeneous data sources. When a data interoperability requester submits a query request, it first performs query parsing, transforming the user's high-level query statement into a logical query tree within the engine. Subsequently, it performs intelligent query rewriting and optimization. This includes: identifying the heterogeneous data sources involved in the query; decomposing and translating the logical query into native query statements (such as SQL, CQL, MongoDB Query Language, etc.) executable by each underlying data source based on the mapping relationship provided by the preset logical data model; and performing distributed query optimization. The optimization strategy considers multiple factors, such as data distribution, data source processing capabilities, network transmission overhead, index availability, and data locality. For example, it intelligently selects which data source to perform filtering, projection, or aggregation operations on to minimize data transfer volume; determines the best join strategy, such as pulling data to the engine layer for in-memory joins, or pushing some join operations to supported underlying data sources for execution; and utilizes parallel computing capabilities to send subqueries to multiple data sources simultaneously and collect results. Furthermore, this application supports advanced optimization techniques such as query caching, materialized views, and pre-computation to accelerate highly repetitive or computationally intensive queries. These intelligent optimizations significantly reduce latency in cross-source queries, improve the real-time performance of big data analytics, and enable users to quickly extract insights from complex, heterogeneous data, thereby supporting more agile business decisions.

[0032] In big data interoperability scenarios, data often involves multiple sensitive sources and high-value information; therefore, security and privacy are inviolable cornerstones. To ensure the confidentiality, integrity, and availability of all data throughout its entire lifecycle (from collection, transmission, storage to use and destruction), and to strictly comply with relevant privacy regulations, this embodiment, after performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instruction, further includes: analyzing the target data corresponding to the data interoperability to determine whether the target data contains sensitive data; the sensitive data is data that the data interoperability requester is prohibited from accessing in advance; if the target data contains the sensitive data, then performing data desensitization on the sensitive data based on a data desensitization method to obtain the desensitized target data; the data desensitization method includes data masking, data generalization, data truncation, data perturbation, and cryptographic hashing. Specifically, At the data transmission level, this application employs end-to-end encryption technologies (such as TLS / SSL) to ensure secure data transmission and prevent eavesdropping or tampering during transmission. For data storage, this application supports static data encryption, encrypting data stored in a distributed file system or database, ensuring that the data cannot be directly read even if the storage medium is illegally accessed. Secondly, access control is a core function. Role-based or attribute-based access control mechanisms will be implemented to finely manage access permissions for different users or system components. Only authorized users can access specific data, and their operation permissions (read, write, modify, delete) will be strictly limited. This includes managing access credentials for data sources, ensuring that only the engine itself can access external data sources with limited permissions. For privacy-sensitive data, this application provides various de-identification methods, such as: data anonymization (removing or obfuscating personally identifiable information), data pseudonymization (replacing real identities with pseudonyms), data generalization (replacing specific data with broader categories), and data perturbation (introducing random noise to protect privacy). These technologies can protect user privacy to the greatest extent possible without affecting data analysis results. This application also provides detailed audit logging capabilities, recording all data access and operations to facilitate security audits and compliance reviews. Through these comprehensive security and privacy protection measures, a trustworthy data environment is built, ensuring the security of data assets and helping enterprises meet increasingly stringent data privacy regulations.

[0033] More specifically, regarding data transmission, all network communications are enforced to use strong encryption protocols to prevent data from being eavesdropped on or tampered with during transmission. For statically stored data, transparent data encryption is supported. By encrypting data in HDFS files and NoSQL databases, the data content cannot be directly interpreted even if the storage medium is accessed without authorization. The core security mechanism is fine-grained access control, which implements minimum access permissions for data resources based on multiple dimensions such as user roles, permission groups, data tags, and operation types. For example, different users can only access their authorized subset of data and can only perform permitted operations (query, update, delete, etc.). This application also integrates identity authentication and authorization services, supporting integration with existing enterprise authentication systems such as LDAP, OAuth2, or Kerberos to achieve single sign-on and centralized access control. For sensitive data, this application provides various data desensitization and anonymization techniques, including data masking, data generalization, data truncation, data perturbation, and cryptographic hashing, to maximize the protection of personal privacy information while meeting analytical needs, complying with data privacy regulations such as GDPR and CCPA. Furthermore, this application features comprehensive security auditing and logging capabilities, meticulously recording all data access behaviors, permission changes, and anomalies, providing a basis for security incident tracing and compliance auditing. Through these comprehensive security measures, this application provides a reliable security barrier for intelligent heterogeneous data interoperability engines.

[0034] In this embodiment, after constructing a preset logical data model based on the target entity, the method further includes: correspondingly, after performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instructions, the method further includes: visualizing the data access results corresponding to the data interoperability to generate a corresponding data visualization table, and displaying the data visualization table on the preset visualization interface. Specifically, an intuitive data source configuration interface is provided, allowing users to easily add, edit, or delete heterogeneous data sources through a graphical wizard without writing any code. This application automatically acquires and displays the metadata of the data source to help users understand the data structure. Regarding data mapping definition, users can establish mapping relationships from different data source fields to unified logical model fields on a visual canvas through dragging, clicking, etc., and users can directly confirm or fine-tune them. The interface also supports complex transformation function configuration to meet more advanced data transformation needs. In addition, this application provides an intelligent query builder, allowing users to construct cross-source queries through graphical operations such as selecting tables, fields, setting filter conditions, and joining relationships, and also supports direct input in the SQL query editor. Query results can be displayed in tables, lists, or various data visualization charts (such as bar charts, line charts, pie charts, scatter plots, maps, etc.). Users can interactively manipulate the charts, such as drill-down, filtering, and zooming, to gain deeper insights into the data. This application also includes dashboard building functionality, allowing users to create custom, real-time, multi-dimensional data views. Through a highly user-friendly interactive design, this application enables data scientists, business analysts, and even ordinary business users to easily and quickly explore and utilize heterogeneous big data, thereby improving the efficiency and intelligence of the entire data analysis and decision-making process.

[0035] Next, a specific operation will be used to illustrate the technical solution in this application.

[0036] Suppose a large retail enterprise faces a severe data silo challenge: its online e-commerce platform's data is stored in a NoSQL database (such as MongoDB), containing user browsing history, shopping cart information, and order details; offline physical store sales data and membership information are stored in traditional relational databases (such as MySQL); and marketing campaign feedback data (such as click-through rates and conversion rates) may come from third-party marketing platform APIs or log files stored on HDFS; furthermore, customer service center call logs are stored in unstructured text format. The enterprise's goal is to establish a unified 360-degree view of its customers to enable omnichannel customer behavior analysis, precision marketing, and personalized recommendations.

[0037] In this scenario, the operational flow of this application is as follows: First, connectors are configured to connect to MongoDB, MySQL, a third-party marketing platform API, and the HDFS file system. For MongoDB, a dedicated connector efficiently retrieves online user behavior logs and transaction data in JSON format. For MySQL, a JDBC connector is used to obtain detailed information about offline members and store transaction records, and the relational data is converted into a unified table structure within the engine according to preset rules. For the marketing API, this application periodically calls the API interface to obtain marketing activity feedback data in JSON format and parses and adapts it. For customer service call record text files stored on HDFS, this application uses a built-in text parser to extract and structure unstructured text content, such as identifying keywords and sentiment tags. During data transmission, all connections are encrypted using TLS / SSL to ensure secure data transmission. This application also performs preliminary format verification and cleaning on the collected data, such as removing duplicate records and handling missing values, to ensure data quality.

[0038] Next, all collected raw data and metadata will be automatically analyzed from online user IDs in MongoDB, member IDs in MySQL, and user identifiers in API data. Machine learning algorithms (such as entity matching and fuzzy matching) will be used to intelligently identify that these user identifiers from different sources actually point to the same "customer" entity. By analyzing the data content, naming conventions, and contextual information of these fields, a unified "customer ID" field can be automatically recommended. Similarly, it will identify that the "product name" of online orders and the "product name" of offline sales records are the same concept and suggest mapping them to a unified "product name" field. In addition, the text content of customer service call records will be analyzed, and customer pain points and preference keywords will be extracted using natural language processing technology and mapped to a unified customer tagging system. Through these intelligent identifications, this application will construct a global, unified customer logical model, including multiple dimensions such as basic customer information, online behavior, offline purchases, marketing interactions, and customer service records, and automatically generate complex mapping rules, such as merging the "last name" and "first name" in MySQL and mapping them to the "full name of the customer" in the unified logical model.

[0039] Data that has undergone pattern recognition and mapping will be intelligently stored in different components based on its characteristics. High-frequency online user real-time behavior data (such as browsing history and shopping carts) may be stored in a distributed NoSQL database (such as HBase) to support low-latency, fast queries. Relatively stable structured data, such as transaction history and member profiles, may be stored in a distributed relational database (such as POSGRSQL) or Parquet / ORC format files on HDFS to support efficient batch analysis. Structured summaries and keywords of customer service call records may be stored in a distributed indexing system that supports full-text search (such as Elasticsearch). This application will select the most suitable storage format and compression algorithm for each data type and automatically manage data replicas and shards to ensure high availability and load balancing. All metadata, including data source connection information, data schema, mapping rules, and security policies, will be centrally managed and indexed for convenient and fast retrieval and management.

[0040] Next, the retail company's business analyst wants to analyze Customer Lifetime Value (CLV) across different channels. The analyst submits a cross-source SQL query through a pre-defined visualization page, for example: "Query members who have purchased goods on e-commerce platforms and made more than three purchases in offline stores within the past year, including their average spending, most frequently purchased product categories, and, combined with customer service records, their main complaint types." Upon receiving this complex query, this application will parse it. It will first identify that the query involves basic customer information, online purchases, offline purchases, and customer service records. Then, based on previously established mapping relationships, this application will automatically translate the SQL query into queries for MongoDB, MySQL, and HDFS or Elasticsearch. During the optimization phase, considering data locality, for example, it first filters out member IDs that meet the offline purchase criteria in the MySQL database and pushes the results to the engine layer. Simultaneously, the engine will send subqueries to MongoDB and Elasticsearch in parallel to retrieve online behavior and customer service records. This application intelligently selects the optimal join strategy. For example, it pulls a small amount of filtered data into the engine's memory for efficient hash join, avoiding large-scale data transfer. If the query involves aggregation operations (such as average spending), this application attempts to push some aggregations down to the data source for execution, reducing the amount of data transferred. Furthermore, if the query is high-frequency, the engine may cache its results so that subsequent requests can receive them instantly. Through this intelligent optimization, analysts can obtain customer insight reports across all channels in seconds, which previously might have required hours or even days of manual data integration.

[0041] Throughout the data interoperability process, data security and privacy protection measures are continuously implemented. For example, when an analyst queries customer information, this application rigorously verifies their role and permissions. If an analyst is not authorized to view a customer's detailed address or phone number, this application automatically anonymizes these sensitive fields, ensuring that the analyst can only see information they are authorized to access. All data access and operations are meticulously logged in audit logs for security audits and compliance checks. Even if data resides in distributed storage, it may be encrypted to prevent unauthorized physical access.

[0042] Throughout the process, users can view the system's currently enabled privacy protection level, the types of data being processed, and how their settings affect the system's behavior at any time. If they change their mind, for example, deciding to no longer allow the system to access voice data, they can disable the corresponding option in this application. The system will immediately stop collecting voice data and update its processing strategy, no longer relying on voice data for analysis. This dynamic, real-time configuration capability further strengthens user control and improves system transparency and user trust.

[0043] Finally, the processed results are returned to the visualization page. Analysts can see average customer spending, purchase trends distributed by channel, and common complaint types displayed through word clouds on an intuitive dashboard. They can further explore the behavioral patterns of specific customer groups through interactive operations such as filtering and drill-down. For example, clicking on a complaint type allows them to view all related customer IDs and customer service record summaries. This highly visualized presentation enables businesses to intuitively understand customer needs, discover market opportunities, and develop more precise marketing strategies, such as pushing exclusive in-store offers to members who are active online but don't spend enough offline, or optimizing the supply chain or service processes for products with many complaints.

[0044] As can be seen, in this embodiment, a preset connector is used to connect with the heterogeneous databases of the target retail enterprise and read the operational data from each heterogeneous database to obtain target heterogeneous data. The heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise. Preset artificial intelligence technology is used to perform semantic analysis and data association analysis on the target heterogeneous data to determine the target entities corresponding to each target heterogeneous data, and a preset logical data model is constructed based on the target entities. The preset logical data model stores the data mapping relationship between the target entities and the target heterogeneous data. Information interoperability requests sent by data interoperability requesters to the target retail enterprise are obtained, and based on the logical data model, the information interoperability requests are converted into target interoperability instructions corresponding to each heterogeneous database, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise. That is, this application uses a preset connector to connect with the heterogeneous databases corresponding to the target retail enterprise to read the operational data from each heterogeneous database, then uses preset artificial intelligence technology to analyze the data association relationship between the operational data, and then constructs a preset logical data model. During data interoperability, the obtained information interoperability requests can be transformed into target interoperability instructions corresponding to each heterogeneous database through a pre-defined logical data model, and then the corresponding target interoperability instructions can be executed. This ensures seamless compatibility with diverse data sources during data interoperability. Based on this, through intelligent data pattern recognition and relationship mapping, and utilizing machine learning and artificial intelligence technologies, the bottlenecks of traditional manual integration methods are overcome, achieving automated understanding and fusion of data semantics and structure. This not only significantly reduces data integration costs but also improves the accuracy and adaptability of data integration.

[0045] refer to Figure 2 The present application also discloses an interoperability device for heterogeneous data in retail enterprises, applied to a preset heterogeneous data interoperability engine, comprising: The data reading module 11 is used to connect with the heterogeneous databases of the target retail enterprise using a preset connector, and read the operational data in each of the heterogeneous databases to obtain the target heterogeneous data; the heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise. Data analysis module 12 is used to perform semantic analysis and data association analysis on each of the target heterogeneous data using preset artificial intelligence technology, so as to determine the target entity corresponding to each of the target heterogeneous data, and construct a preset logical data model based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data; The data interoperability module 13 is used to acquire the information interoperability request sent by the data interoperability requester to the target retail enterprise, and convert the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

[0046] As can be seen, in this embodiment, operational data is read from various heterogeneous databases corresponding to the target retail enterprise by using a preset connector. Then, preset artificial intelligence technology is used to analyze the data relationships between these operational data, and a preset logical data model is constructed. During data interoperability, the preset logical data model can be used to transform the obtained information interoperability requests into target interoperability instructions corresponding to each heterogeneous database, and then execute the corresponding target interoperability instructions. This ensures seamless compatibility with diverse data sources during data interoperability. Based on this, through intelligent data pattern recognition and relationship mapping, machine learning and artificial intelligence technologies are used to overcome the bottlenecks of traditional manual integration methods, achieving automated understanding and fusion of data semantics and structure. This not only significantly reduces data integration costs but also improves the accuracy and adaptability of data integration.

[0047] In some specific embodiments, the data reading module 11 may specifically include: The database connection unit is used to determine the preset connectors for each heterogeneous database of the target retail enterprise based on the database type of each heterogeneous database; the preset connectors establish corresponding connections with each heterogeneous database of the target retail enterprise. The data reading unit is used to read the operational data from each of the heterogeneous databases using the preset connector and based on the preset data reading method to obtain the target heterogeneous data; the preset data reading method includes full synchronization, incremental synchronization and real-time streaming data ingestion.

[0048] In some specific embodiments, the data reading module 11 may specifically include: The data encryption unit is used to connect to the heterogeneous databases of the target retail enterprise using a preset connector, and to read the operational data in each of the heterogeneous databases using a preset data transmission encryption method to obtain the initial heterogeneous data. The data processing unit is used to perform data format conversion, character set encoding unification, data type conversion, and data cleaning on each of the initial heterogeneous data to obtain each target heterogeneous data.

[0049] In some specific embodiments, the data analysis module 12 may specifically include: The semantic analysis unit is used to analyze the semantic associations of text metadata in each of the target heterogeneous data using natural language processing technology to obtain a first analysis result; The data clustering unit is used to analyze the data patterns and data in each of the target heterogeneous data using a clustering algorithm, so as to obtain the dataset association relationship between the target heterogeneous data, and determine the dataset association relationship as the second analysis result; The first model building unit is used to build the target semantic model using knowledge graph construction technology; A mapping relationship determination unit is used to determine the target entities corresponding to each of the target heterogeneous data and the data mapping relationship between each of the target heterogeneous data based on the first analysis result and the second analysis result; The first model construction unit is used to construct a preset logical data model based on the target entity, the data mapping relationship, and the target semantic model.

[0050] In some specific embodiments, the data interoperability module 13 may specifically include: The request acquisition unit is used to acquire information interoperability requests sent by the data interoperability requester to the target retail enterprise based on a preset query interface. A database determination unit is used to parse the information interoperability request using the logical data model in order to determine the target database corresponding to the information interoperability request from each of the heterogeneous databases. The instruction conversion unit is used to convert the information interoperability request into the target interoperability instruction corresponding to the target database based on the data mapping relationship stored in the logical data model.

[0051] In some specific embodiments, the interoperability device for heterogeneous data of retail enterprises may further include: The sensitive data determination module is used to analyze the target data corresponding to the data interoperation to determine whether the target data contains sensitive data; the sensitive data is data that the data interoperation requester is prohibited from accessing in advance. The data desensitization module is used to perform data desensitization operations on the sensitive data based on the data desensitization method if the target data contains the sensitive data, so as to obtain the desensitized target data; the data desensitization method includes data masking, data generalization, data truncation, data perturbation and cryptographic hashing.

[0052] In some specific embodiments, the interoperability device for heterogeneous data of retail enterprises may further include: The first data display module is used to display the data mapping relationship stored in the preset logical data model on the visualization canvas of the preset visualization interface, so that the user can modify and adjust the data mapping relationship on the visualization canvas; The second data display module is used to perform graphical processing on the data access results corresponding to the data interoperation to generate a corresponding data visualization table, and to display the data visualization table on the preset visualization interface.

[0053] Furthermore, embodiments of this application also disclose an electronic device, Figure 3 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0054] Figure 3 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the interoperability method for heterogeneous data in retail enterprises disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0055] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0056] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0057] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the interoperability method for heterogeneous retail enterprise data executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0058] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned interoperability method for heterogeneous data of retail enterprises. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0059] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0060] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0061] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0062] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0063] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for interoperating heterogeneous data in a retail enterprise, characterized in that, Applied to pre-defined heterogeneous data interoperability engines, including: The system connects to the heterogeneous databases of the target retail enterprise using a pre-defined connector and reads the operational data from each of the heterogeneous databases to obtain the target heterogeneous data; the heterogeneous databases are deployed in the operational scenarios corresponding to the target retail enterprise. Using preset artificial intelligence technology, semantic analysis and data association analysis are performed on each of the target heterogeneous data to determine the target entity corresponding to each of the target heterogeneous data, and a preset logical data model is constructed based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data. The system acquires the information interoperability request sent by the data interoperability requester to the target retail enterprise, and transforms the information interoperability request into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

2. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, The process of connecting to the heterogeneous databases of the target retail enterprise using a pre-set connector and reading operational data from each of the heterogeneous databases to obtain target heterogeneous data includes: For each heterogeneous database of the target retail enterprise, a preset connector is determined for each heterogeneous database; the preset connector establishes a corresponding connection with each heterogeneous database of the target retail enterprise. The operation data in each of the heterogeneous databases is read using the preset connector and based on the preset data reading method to obtain the target heterogeneous data; the preset data reading method includes full synchronization, incremental synchronization and real-time streaming data ingestion.

3. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, The process of connecting to the heterogeneous databases of the target retail enterprise using a pre-set connector and reading operational data from each of the heterogeneous databases to obtain target heterogeneous data includes: The system connects to the heterogeneous databases of the target retail enterprise using a pre-defined connector, and reads the operational data from each of the heterogeneous databases using a pre-defined data transmission encryption method to obtain the initial heterogeneous data. The initial heterogeneous data are subjected to data format conversion, character set encoding unification, data type conversion and data cleaning to obtain the target heterogeneous data.

4. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, The step of using preset artificial intelligence technology to perform semantic analysis and data association analysis on each of the target heterogeneous data to determine the target entities corresponding to each of the target heterogeneous data, and constructing a preset logical data model based on the target entities, includes: Natural language processing techniques are used to analyze the semantic relationships of text metadata in each of the target heterogeneous data to obtain the first analysis result; Clustering algorithms are used to analyze the data patterns and data in each of the target heterogeneous data sets to obtain the dataset association relationship between the target heterogeneous data sets, and the dataset association relationship is determined as the second analysis result; Utilize knowledge graph construction techniques to build a target semantic model; Based on the first analysis result and the second analysis result, the target entities corresponding to each of the target heterogeneous data and the data mapping relationship between each of the target heterogeneous data are determined; A preset logical data model is constructed based on the target entity, the data mapping relationship, and the target semantic model.

5. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, The process of acquiring data interoperability requests from the requesting party in response to the target retail enterprise, and converting the information interoperability requests into target interoperability instructions corresponding to each of the heterogeneous databases based on the logical data model, includes: Based on the preset query interface, the data interoperability requester obtains the information interoperability request sent by the data interoperability requester to the target retail enterprise. The logical data model is used to parse the information interoperability request in order to determine the target database corresponding to the information interoperability request from each of the heterogeneous databases; Based on the data mapping relationships stored in the logical data model, the information interoperability request is transformed into the target interoperability instruction corresponding to the target database.

6. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, After performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instructions, the method further includes: The target data corresponding to the data interoperation is analyzed to determine whether the target data contains sensitive data; the sensitive data is data that the data interoperation requester is prohibited from accessing in advance. If the target data contains the sensitive data, then the sensitive data is desensitized using a data desensitization method to obtain the desensitized target data; the data desensitization method includes data masking, data generalization, data truncation, data perturbation, and cryptographic hashing.

7. The interoperability method for heterogeneous data in retail enterprises according to claim 1, characterized in that, After constructing the preset logical data model based on the target entity, the method further includes: The data mapping relationship stored in the preset logical data model is displayed on the visualization canvas of the preset visualization interface, so that the user can modify and adjust the data mapping relationship on the visualization canvas; Accordingly, after performing data interoperability on the operational data in each of the heterogeneous databases based on the target interoperability instructions, the method further includes: The data access results corresponding to the data interoperation are processed into images to generate corresponding data visualization tables, and the data visualization tables are displayed on the preset visualization interface.

8. An interoperability device for heterogeneous data in a retail enterprise, characterized in that, Applied to pre-defined heterogeneous data interoperability engines, including: The data reading module is used to connect with the heterogeneous databases of the target retail enterprise using a preset connector, and read the operational data in each of the heterogeneous databases to obtain the target heterogeneous data; the heterogeneous databases are deployed in the corresponding operational scenarios of the target retail enterprise. The data analysis module is used to perform semantic analysis and data association analysis on each of the target heterogeneous data using preset artificial intelligence technology, so as to determine the target entity corresponding to each of the target heterogeneous data, and construct a preset logical data model based on the target entity; the preset logical data model stores the data mapping relationship between the target entity and each of the target heterogeneous data. The data interoperability module is used to acquire information interoperability requests sent by the data interoperability requester to the target retail enterprise, and to convert the information interoperability requests into target interoperability instructions corresponding to each heterogeneous database based on the logical data model, so as to perform data interoperability on the operational data in each heterogeneous database based on the target interoperability instructions, and obtain the corresponding heterogeneous data interoperability results of the target retail enterprise.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the interoperability method for heterogeneous data in retail enterprises as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer programs, which, when executed by a processor, implement the interoperability method for heterogeneous data of retail enterprises as described in any one of claims 1 to 7.

Images