Search

CN-121996207-A - Intelligent automatic analysis and integrated fusion processing system for multi-source data

CN121996207ACN 121996207 ACN121996207 ACN 121996207ACN-121996207-A

Abstract

The invention discloses an intelligent automatic analysis and integration fusion processing system for multi-source data, and belongs to the field of software automation. The system comprises a multi-source input unit, a unified semantic modeling unit, an automatic analysis unit, an automatic preprocessing unit, an automatic cooperative execution unit and a security authentication mechanism, wherein the security authentication mechanism comprises primary security authentication and secondary security authentication. The invention can realize end-to-end automatic closed loop from the understanding of the requirement to the automatic analysis and processing of the path to the execution of the landing, and accurately respond to the key requirements in the complex project service scene.

Inventors

  • YANG WEIDONG

Assignees

  • 法正智能科技有限公司

Dates

Publication Date
20260508
Application Date
20251124

Claims (8)

  1. 1. An intelligent automated analysis and integration fusion processing system for multi-source data, comprising: the multi-source input unit is used for respectively receiving the multi-source heterogeneous data document uploaded by the user and the input natural language; The unified semantic modeling unit is connected with the multi-source input unit and is used for performing semantic decoupling on the multi-source content after file identification and extraction, performing intelligent reconstruction and normalized modeling, and automatically archiving and warehousing; The automatic analysis unit is connected with the unified semantic modeling unit, and is used for disassembling intention elements input by a user through the multi-layer semantic analysis engine, identifying user behavior targets, action objects, constraint basis and implicit subtasks, constructing dynamic semantic association graphs, associating project documents, frame assets and user history information to fill semantic gaps, intelligently searching paths based on semantic integrity, deducing supported query dimensions, input parameter formats, return structure norms and business rules and format constraints to be followed when data are injected; The automatic preprocessing unit is connected with the automatic analysis unit and intelligently generates an expected execution scheme which accords with an initial target of a user; The automatic collaborative execution unit is connected with the automatic preprocessing unit, and the unit starts to execute after the user confirms no error according to the expected execution scheme; The security authentication mechanism comprises a first security authentication and a second security authentication and is used for realizing the whole-process security control and intention alignment.
  2. 2. The intelligent automated analysis and integration fusion processing system for multi-source data according to claim 1, wherein the multi-source input unit comprises an automatic identification of input source type, a lightweight pre-check of multi-mode data to filter invalid or damaged inputs, and a unified aggregation of user input requirements or problem descriptions and metadata into a buffer pool.
  3. 3. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 1, wherein the multi-source input unit is used for carrying out format recognition on uploaded multi-source heterogeneous files, extracting file content information and carrying out pre-verification.
  4. 4. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 1, wherein the unified semantic modeling unit comprises the following specific processes: preprocessing the multi-source heterogeneous file, stripping format differences, and restoring the multi-source heterogeneous file into a uniform intermediate representation form; Extracting key entities, relationships among the entities, operation intents, context constraints and implicit logic rules based on the deep semantic understanding of the domain ontology; performing semantic decoupling, and restoring the original expression into an atomization semantic unit of a main body-action-target-constraint; performing intelligent completion and logic reconstruction, and obtaining potential functional module division, data flow paths or dependency relationships through pattern recognition to generate standardized expression conforming to a preset semantic paradigm; coding and classifying the reconstructed semantic units in a hierarchical mode to form a structured semantic object set with rich meta semantic tags.
  5. 5. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 1, wherein the automatic preprocessing unit comprises the following specific processes: Based on the structured execution intention, automatically deducing a plurality of execution paths capable of achieving the target; each execution path is independently previewed, and a request example, a simulation response structure, estimated execution time consumption and a dependent service state are generated; applying a rule engine and a permission policy to carry out compliance and risk assessment on each path; presenting all the previewing schemes to a user in a structured, comparative and interactive manner; According to the selection of the user, a complete, traceable and revisable expected execution scheme is output.
  6. 6. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 4, wherein the automatic cooperative execution unit comprises the following specific processes: analyzing the structured execution scheme confirmed by the user, and extracting task sequences, dependent resources, target frames, parameter contracts and execution constraints; Automatically generating an executable instruction stream according to the execution intention path output by the automatic analysis unit, and dynamically matching the adapter according to the type of the target carrier; Intelligently arranging and scheduling according to the task dependency relationship, and realizing cooperative execution across multiple platforms or tools; The state, time consumption and abnormality of each subtask are monitored in real time, a visual execution track map is constructed, and self-recovery processing is carried out according to a preset strategy; after the task is completed, the output objects of all links are collected, structured and packaged into an execution result package and archived.
  7. 7. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 1, wherein the first security authentication comprises the following specific processes: Performing source validity verification on original materials uploaded by users, and identifying and blocking unauthorized formats and malicious files; Performing sensitive word recognition, violation intention detection and context risk assessment on the text, the image OCR result and the structured field description through the compliance semantic model; Judging whether the input content exceeds the operable range of the roles by combining the user identity, the project authority and the organization strategy; Isolating the identified risk content and generating a risk alert immediately informing the user that only content that passes verification is allowed to enter subsequent processing.
  8. 8. The intelligent automatic analysis and integration fusion processing system for multi-source data according to claim 1, wherein the secondary security authentication comprises the following specific processes: Comparing the pre-generated execution scheme with the original intention of the user, and detecting semantic deviation, excessive deduction or missing constraint; Carrying out grammar security verification, data compliance check and behavior validity judgment on an instruction to be executed; if the risk is detected, automatically intercepting the execution flow and feeding back risk details and modification suggestions to the user; The automated collaborative execution unit is authorized to initiate a real operation only after final confirmation by the user.

Description

Intelligent automatic analysis and integrated fusion processing system for multi-source data The application is the following application Application number 2025117245080 The filing date is 2025.11.24 Application name is a divisional application of an automatic multisource analysis and integrated processing system based on artificial intelligence Technical Field The invention relates to the field of software automation, in particular to an intelligent automatic analysis and integration fusion processing system for multi-source data. Background In the field of software automation, conventional methods typically rely on specialized artificial intelligence models to perform input and output of a single task in isolation, e.g., based solely on interface document descriptions, construct simulation data using generative artificial intelligence and perform calls. Meanwhile, an automatic RPA tool existing in the market also depends on manual programming or depends on an artificial intelligence natural language processing model to generate a script program after specified understanding. The two modes also have complex operation, require a large amount of manual arrangement and specific prompt words for construction, require manual intervention to understand context, manually arrange interface calling sequences, manually preset database basic data, and even maintain parameter mapping and state transfer item by item. The method not only leads to fragile and difficult maintenance of an automatic link, but also is easier to miss hidden dependence of key business constraint, set parameters, boundary conditions and the like, especially under the scene that business complexity is increased and parameter combination is exponentially increased, if multi-source heterogeneous project data such as comprehensive requirement documents, design description, data models and the like are required to be cooperatively compiled, the traditional mode exposes serious efficiency bottlenecks, the maintenance cost is increased suddenly, the coverage scene is fragmented and the artificial omission frequently occurs, and finally, the automation is enabled to be semi-automatic, so that the large-scale, high-fidelity and self-adaptive engineering practice demands are difficult to support. When the real complex business scene on the project is faced and the business logic, data dependence and flow relation in the requirement document are combined, the automatic serial connection of multiple interfaces, parameter linkage and closed-loop execution are realized, the existing conditions are low in efficiency and limited in effect. Therefore, how to provide an automated multi-source analysis and integrated processing system based on artificial intelligence is a problem to be solved by those skilled in the art. Disclosure of Invention In view of the above, the invention provides an intelligent automatic analysis and integration fusion processing system for multi-source data, which realizes end-to-end automatic closed loop from the understanding of the needs to the automatic analysis processing path to the execution of the landing, and accurately responds to the key requirements in the complex project service scene so as to solve the technical problems in the prior art. In order to achieve the above purpose, the present invention adopts the following technical scheme: An intelligent automated analysis and integration fusion processing system for multi-source data, comprising: the multi-source input unit is used for respectively receiving the multi-source heterogeneous data document uploaded by the user and the input natural language; The unified semantic modeling unit is connected with the multi-source input unit and is used for performing semantic decoupling on the multi-source content after file identification and extraction, performing intelligent reconstruction and normalized modeling, and automatically archiving and warehousing; The automatic analysis unit is connected with the unified semantic modeling unit, and is used for disassembling intention elements input by a user through the multi-layer semantic analysis engine, identifying user behavior targets, action objects, constraint basis and implicit subtasks, constructing dynamic semantic association graphs, associating project documents, frame assets and user history information to fill semantic gaps, intelligently searching paths based on semantic integrity, deducing supported query dimensions, input parameter formats, return structure norms and business rules and format constraints to be followed when data are injected; The automatic preprocessing unit is connected with the automatic analysis unit and intelligently generates an expected execution scheme which accords with an initial target of a user; The automatic collaborative execution unit is connected with the automatic preprocessing unit, and the unit starts to execute after the user confirms no error according to the expected execution s