Enterprise-level information management platform and method based on multi-source data fusion

Abstract

The application discloses an enterprise-level information management platform and method based on multi-source data fusion, belongs to the technical field of enterprise data management and information processing, and aims at the problems of data islands, insufficient consistency of multi-source data fusion, missing data traceability links, insufficient granularity of permission control, and disconnection between data and business execution in existing enterprise information management. The application first completes multi-source data standardization preprocessing with a business domain label and data blood relationship identification, then generates a global fusion data set through hierarchical fusion and conflict correction, completes hierarchical permission control and data visualization distribution in combination with an enterprise organization framework, realizes closed-loop updating of the fusion data through business incremental data backflow, can effectively break through enterprise data islands, improves data availability and information management efficiency, and is suitable for operation and management requirements of enterprise multi-business collaboration.

Description

Technical Field

[0001] This invention relates to the field of enterprise data management and information processing technology, and in particular to an enterprise-level information management platform and method based on multi-source data fusion. Background Technology

[0002] As enterprises continue their digital transformation, most have gradually deployed multiple business systems such as ERP, CRM, OA, and production management. Simultaneously, they connect to external data sources such as supply chains and industry regulations. The scale and types of data both internally and externally are continuously growing, making enterprise-level information management a core component supporting business operations and management decisions. Currently, most conventional enterprise information management solutions focus on controlling information from single business systems. While some solutions achieve data aggregation and collection from multiple systems, they lack systematic data fusion and full lifecycle management capabilities, making it difficult to adapt to the management needs of collaborative operations across multiple business lines.

[0003] Most existing enterprise multi-source data management technologies only perform simple data aggregation and format conversion, failing to establish a hierarchical fusion mechanism tailored to the characteristics of enterprise business domains. This makes it difficult to resolve cross-system data correlation and conflict issues, easily leading to insufficient consistency and availability of the merged data. Furthermore, most solutions lack complete data lineage identification and traceability links, making it difficult to meet enterprise data compliance audit requirements. Data access control is primarily at the system or module level, unable to achieve fine-grained field-level access control configuration, increasing the risk of data leakage or access violations. In addition, existing solutions often fail to deeply integrate data management with business process execution, lacking a closed-loop feedback and update mechanism for incremental data, resulting in a disconnect between management data and actual business conditions, and failing to provide continuous and real-time information support for enterprise business operations. Summary of the Invention

[0004] To address the aforementioned issues, this application provides an enterprise-level information management platform and method based on multi-source data fusion, enabling standardized integration, end-to-end control, and closed-loop iteration of enterprise multi-source data, thereby improving the overall efficiency and data availability of enterprise information management.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: an enterprise-level information management method based on multi-source data fusion, comprising the following steps:

[0006] S1 connects to multiple sources of business data from within and outside the enterprise, cleans and standardizes the format of the raw data, and simultaneously tags the preprocessed data with business domain labels and data lineage identifiers to generate standardized datasets.

[0007] S2 performs hierarchical fusion processing on standardized datasets based on business domain labels. First, it completes the association and fusion of data within the same business domain, and then completes the feature matching and fusion of cross-business domain data to generate a full-domain fused dataset. During the fusion process, data conflicts are automatically verified, and weights are assigned and corrected according to preset source priority and update time, while retaining the data traceability link throughout the entire process.

[0008] Based on the enterprise's organizational structure and job roles, S3 configures hierarchical access permissions for the full-domain fusion dataset. At the same time, according to the needs of different business scenarios, it generates corresponding visual management dashboards with matching permissions and distributes the fusion data within the permission scope to the corresponding business ports.

[0009] S4 connects to the execution nodes of various business processes of an enterprise based on distributed fusion data, collects incremental data generated by business execution in real time, and feeds the incremental data back to the S1 step to complete preprocessing, thereby realizing closed-loop update and iteration of the fusion dataset.

[0010] Furthermore, in step S1, the multi-source business data from both internal and external sources include data from the enterprise's internal ERP, CRM, OA, and production management systems, as well as publicly available data from external supply chains and industry regulators. Data access adopts an access method that matches the source type, including but not limited to API interface integration, direct database connection, and batch file import.

[0011] Furthermore, in step S1, the format standardization preprocessing is based on the pre-established unified enterprise data element standard, which completes field mapping, unit unification and format conversion for data of the same type, extracts key information from unstructured data and converts it into structured tag data, and performs desensitization and encryption processing on sensitive fields.

[0012] Furthermore, in step S2, the preset source priority is pre-classified based on the system authority, update frequency and business relevance of the data source. Data with higher priority corresponds to a higher weight ratio when correcting conflicts.

[0013] Furthermore, in step S2, the data association and fusion within the same business domain uses the unique identifier of the business entity as the association primary key to complete the splicing of multi-dimensional data of the same entity; the feature matching and fusion across business domains is based on preset business association rules to complete the association mapping of data from different business domains.

[0014] Furthermore, in step S3, the hierarchical access permissions follow the principle of least privilege, allowing for the configuration of access, editing, and export permissions for individual data fields. At the same time, all data operation behaviors are logged, supporting the traceability and auditing of operation behaviors.

[0015] Furthermore, in step S3, the visual management dashboard supports custom field filtering, chart configuration, and report generation within permissions. It can also configure data warning thresholds, and automatically push warning information to the corresponding person in charge when the data exceeds the threshold range.

[0016] Furthermore, in step S4, the incremental data reflow supports two modes: fixed-period batch reflow and real-time reflow triggered by business nodes. After the reflowed data completes preprocessing, it is automatically updated to the global fusion dataset and the corresponding data lineage is updated synchronously.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages compared with the prior art:

[0018] This invention effectively breaks down data silos between different business systems and data sources within an enterprise. Through standardized preprocessing and hierarchical fusion mechanisms, it enhances the consistency and availability of multi-source heterogeneous data fusion. By retaining data lineage identification and traceability links throughout the entire process, it adapts to the management needs of enterprise data compliance auditing. Through refined hierarchical access control, it reduces the risk of data leakage and unauthorized access, adapting to the management needs of different levels within the enterprise. Furthermore, through a closed-loop incremental data feedback and update mechanism, it achieves deep integration of management data with the business execution process, ensuring the timeliness of management data. This provides stable and reliable information support for enterprise business operations and management decisions, effectively improving the overall efficiency and business adaptability of enterprise-level information management.

[0019] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0020] Figure 1 This is a sequence diagram of the overall enterprise-level information management method of the present invention;

[0021] Figure 2 This is a flowchart of the multi-source data access and standardized preprocessing process of the present invention;

[0022] Figure 3 This is a schematic diagram of the hierarchical fusion processing and conflict resolution mechanism of the present invention;

[0023] Figure 4 This is a timing diagram of the data closed-loop update and iteration process of the present invention;

[0024] Figure 5 This is a diagram of the multi-source data access method and data governance architecture of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] It should be noted that the terms "vertical," "horizontal," "up," "down," "left," "right," and similar expressions used in this article are for illustrative purposes only and do not represent the only possible implementation.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention; the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] The following is in conjunction with the appendix Figure 1-5 The invention will be further explained through specific embodiments.

[0029] In one optional implementation scenario, this method is applied to the enterprise-level information management process of a medium-sized manufacturing enterprise. First, it connects with the enterprise's internal ERP, CRM, OA, and production execution systems, integrating business data. Simultaneously, it connects with upstream suppliers' supply chain management systems and publicly available compliance data from industry regulatory authorities. Depending on the data source type, it employs API interface integration, direct database connection, and batch file import to complete data access. For the raw data, based on the enterprise's pre-defined unified data element standards, preprocessing operations are performed, including data cleaning, null value imputation, field format standardization, and unit conversion. Key information is extracted from unstructured data such as contract texts and product descriptions and converted into structured tags. Sensitive fields such as customer contact information and financial data are anonymized and encrypted. Each preprocessed data entry is tagged with corresponding business domain labels such as purchasing, production, sales, and finance, as well as a lineage identifier recording the data source and update time, generating a standardized dataset.

[0030] Subsequently, the standardized dataset is hierarchically fused based on business domain labels. First, using unique identifiers such as supplier codes, material codes, and customer codes as primary keys, data associations within the same business domain, such as procurement, production, sales, and finance, are completed. Then, based on preset business association rules, such as the association rules between material codes and production orders and sales orders, feature matching and fusion of cross-business domain data are completed to generate an enterprise-wide fused dataset. During the fusion process, conflicting data of the same entity are automatically verified. According to the authority and update frequency of the data source system, a pre-set priority is applied, and weights are assigned and corrected based on the data update time. Data with higher priority and more recent update time is given priority. At the same time, the entire data fusion traceability chain is preserved, recording the source and fusion correction process of each piece of data.

[0031] After data fusion is completed, hierarchical access permissions are configured for the fused dataset based on the enterprise's organizational structure and job roles. Access and editing permissions for corresponding business domains and data fields are configured for different positions such as front-line production staff, department managers, and enterprise decision-makers, following the principle of least privilege. At the same time, all data viewing, editing, and export operations are fully logged to support traceability and auditing of operational behavior. For different business scenarios such as production management, sales management, financial management, and procurement management, visual management dashboards matching the permissions of corresponding positions are generated, supporting field filtering, chart configuration, and automatic generation of operational reports within the permissions. Corresponding early warning thresholds are configured for inventory data, order delivery data, and financial collection data. When the data exceeds the threshold range, early warning information can be automatically pushed to the corresponding business person in charge.

[0032] Finally, the fused data with matching permissions is distributed to various business ports of the enterprise, connecting to the execution nodes of business processes such as procurement, production, sales, and finance. Incremental data such as procurement receipts, production reports, order signings, and financial payments generated during business execution are collected in real time. Depending on the needs of the business scenario, two modes are adopted: real-time backflow triggered by business nodes and batch backflow with fixed periods. The incremental data is backflowed to the data preprocessing stage, and after the incremental data is standardized, it is automatically updated to the full-domain fused dataset. The corresponding data lineage is updated synchronously, realizing the closed-loop update and continuous iteration of the enterprise's fused dataset, providing continuous information support for the enterprise's daily operation management and decision-making.

[0033] The enterprise-level information management platform based on multi-source data fusion, as described in this invention, adopts a layered, loosely coupled architecture design. The overall framework is divided into a five-layer, integrated system, namely, the basic support layer, the data access and preprocessing layer, the multi-source data hierarchical fusion layer, the business control and distribution layer, the scenario-based application layer, and the security control and auditing system that runs through all layers of the architecture. The components at each layer interact through standardized interfaces, which can be seamlessly integrated with the enterprise's existing business systems and has flexible expansion capabilities to adapt to different enterprise sizes and business scenarios.

[0034] 1. Basic support layer

[0035] The foundational support layer provides the underlying resources and environment support for the overall operation of the platform. Its core components include server hardware resource clusters, operating system and runtime environment components, distributed database clusters, message middleware components, and containerized deployment components. Among these, the distributed database cluster can simultaneously support the categorized storage of structured, semi-structured, and unstructured data; the message middleware component ensures the stability and integrity of multi-source data transmission and adapts to high-concurrency data acquisition and distribution scenarios; and the containerized deployment component enables elastic scaling of platform resources based on the enterprise's data scale and business concurrency requirements, allowing for functional expansion and resource adaptation without adjusting the overall platform architecture.

[0036] 2. Data Access and Preprocessing Layer

[0037] The data access and preprocessing layer provides the platform with data acquisition and standardized processing capabilities. Corresponding to the data preprocessing steps of the method of this invention, the core components include a multi-source data docking component, a data standardization preprocessing component, and a data tagging and lineage identification component. The multi-source data integration component has built-in multi-type adaptation interfaces, supporting integration with internal data sources such as enterprise ERP, CRM, OA, and production management systems, as well as external data sources such as upstream supply chain, downstream channels, and publicly available industry regulatory data. It can adapt to various access methods such as API interface integration, direct database connection, and batch file import, completing the collection and transmission of full-volume raw data. The data standardization preprocessing component is based on a preset enterprise unified data element standard, completing the cleaning, null value filling, field mapping, unit unification, and format conversion of raw data. At the same time, it can extract key information from unstructured data such as contract text and product descriptions and convert it into structured tagged data, and perform de-identification and encryption processing on sensitive fields such as customer privacy information and core financial data. The data tagging and lineage identification component can mark the preprocessed standardized data with corresponding business domain tags, and simultaneously generate a unique data lineage identifier for each data, fully recording basic information such as the data source system, collection time, and update records, providing basic support for subsequent data fusion and full-chain traceability.

[0038] 3. Multi-source data hierarchical fusion layer

[0039] The multi-source data hierarchical fusion layer is the core data processing layer of the platform, corresponding to the hierarchical fusion steps of the method of this invention. Its core components include a data association fusion component within the same business domain, a cross-business domain feature matching fusion component, a data conflict verification and correction component, and a data traceability link management component. The data association fusion component within the same business domain uses unique identifiers of business entities such as supplier codes, material codes, and customer codes as association keys to complete the association and splicing of multi-source data within the same business domain, integrating multi-dimensional data of the same business entity to form a single-business-domain fusion dataset. The cross-business-domain feature matching fusion component, based on preset business association rules, completes feature matching and association mapping between datasets from different business domains, achieving deep fusion of cross-business-domain data and ultimately generating an enterprise-wide fusion dataset. The data conflict verification and correction component can automatically identify conflicting data of the same entity during the fusion process, assigning weights according to preset source priorities and data update times to correct conflicting data and ensure the consistency of the fused data. The data traceability link management component can retain the entire process operation record of data fusion, the unique lineage identifier of associated data, and construct a complete data traceability link, supporting full-process traceability of data from original collection to fusion output.

[0040] 4. Business Control and Distribution Layer

[0041] The business control and distribution layer is the core control layer of the platform, corresponding to the permission configuration and data distribution steps of the method of this invention. Its core components include a hierarchical permission control component, a visual dashboard configuration component, and a data distribution and early warning component. The hierarchical permission control component, based on the enterprise's organizational structure and job roles, configures granular hierarchical access permissions for the entire integrated dataset. It supports configuring access, editing, and export permissions for individual data fields, adhering to the principle of least privilege to adapt to the management needs of different levels and positions within the enterprise. The visual dashboard configuration component can configure visual management dashboards that match job permissions according to the needs of different business scenarios, providing customization capabilities for field filtering, chart configuration, and report generation. Dashboard adjustments and optimizations can be completed without complex coding operations. The data distribution and early warning component can accurately distribute integrated data within the permission scope to the corresponding business ports. It also supports configuring data early warning thresholds; when data exceeds the threshold range, it automatically pushes early warning information to the corresponding business manager, achieving early prevention of business risks.

[0042] 5. Scenario-based application layer

[0043] The scenario-based application layer serves as the linkage layer between the platform and enterprise business processes, corresponding to the closed-loop update steps of the method in this invention. Its core components include a business process integration component, an incremental data acquisition component, and a data closed-loop iteration component. The business process integration component can connect to the execution nodes of various business processes within the enterprise, such as procurement, production, sales, finance, and human resources, achieving deep linkage between platform-integrated data and business execution processes, providing corresponding information support for frontline business execution, departmental management, and enterprise decision-making. The incremental data acquisition component can collect incremental business data generated during business process execution in real time, supporting both real-time acquisition triggered by business nodes and fixed-period batch acquisition modes, ensuring the completeness and timeliness of incremental data collection. The data closed-loop iteration component can feed the collected incremental data back to the data access and preprocessing layer, completing the standardized processing of the incremental data before synchronously updating it to the enterprise-wide integrated dataset, and simultaneously updating the corresponding data lineage links, achieving closed-loop updates and continuous iteration of the integrated dataset.

[0044] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make various modifications and alterations without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the claims.

Claims

1. An enterprise-level information management method based on multi-source data fusion, characterized in that: Includes the following steps: S1 connects to multiple sources of business data from within and outside the enterprise, cleans and standardizes the format of the raw data, and simultaneously tags the preprocessed data with business domain labels and data lineage identifiers to generate standardized datasets. S2 performs hierarchical fusion processing on standardized datasets based on business domain labels. First, it completes the association and fusion of data within the same business domain, and then completes the feature matching and fusion of cross-business domain data to generate a full-domain fused dataset. During the fusion process, data conflicts are automatically verified, and weights are assigned and corrected according to preset source priority and update time, while retaining the data traceability link throughout the entire process. Based on the enterprise's organizational structure and job roles, S3 configures hierarchical access permissions for the full-domain fusion dataset. At the same time, according to the needs of different business scenarios, it generates corresponding visual management dashboards with matching permissions and distributes the fusion data within the permission scope to the corresponding business ports. S4 connects to the execution nodes of various business processes of an enterprise based on distributed fusion data, collects incremental data generated by business execution in real time, and feeds the incremental data back to the S1 step to complete preprocessing, thereby realizing closed-loop update and iteration of the fusion dataset.

2. The method according to claim 1, characterized in that, In step S1, the multi-source business data from both internal and external sources include data from the enterprise's internal ERP, CRM, OA, and production management systems, as well as publicly available data from external supply chains and industry regulators. Data access adopts an access method that matches the source type, including but not limited to API interface integration, direct database connection, and batch file import.

3. The method according to claim 1, characterized in that, In step S1, the format standardization preprocessing is based on the pre-established unified enterprise data element standard. It completes field mapping, unit unification and format conversion for data of the same type, extracts key information from unstructured data and converts it into structured tagged data, and performs desensitization and encryption processing on sensitive fields.

4. The method according to claim 1, characterized in that, In step S2, the preset source priority is pre-classified based on the system authority, update frequency and business relevance of the data source. Data with higher priority has a higher weight ratio when correcting conflicts.

5. The method according to claim 1, characterized in that, In step S2, data association and fusion within the same business domain uses the unique identifier of the business entity as the association key to complete the splicing of multi-dimensional data of the same entity; feature matching and fusion across business domains is based on preset business association rules to complete the association mapping of data from different business domains.

6. The method according to claim 1, characterized in that, In step S3, the hierarchical access permissions follow the principle of least privilege. Access, editing, and export permissions can be configured for individual data fields. At the same time, all data operation behaviors are logged to support traceability and auditing of operation behaviors.

7. The method according to claim 1, characterized in that, In step S3, the visual management dashboard supports custom field filtering, chart configuration, and report generation within permissions. It can also configure data warning thresholds, and automatically push warning information to the corresponding person in charge when the data exceeds the threshold range.

8. The method according to claim 1, characterized in that, In step S4, incremental data reflow supports two modes: fixed-period batch reflow and real-time reflow triggered by business nodes. After the reflowed data completes preprocessing, it is automatically updated to the global fusion dataset and the corresponding data lineage is updated synchronously.

Images