Method and device for automatically generating and dynamically configuring research and development project nodes
By introducing named entity recognition technology and multi-objective optimization algorithms, the semantic gap problem in the configuration of R&D project nodes was solved, achieving precise alignment and efficient dynamic adjustment of policies and projects, and improving the system's response speed and data collection capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-05-26
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies suffer from semantic gaps when mapping unstructured policy texts to structured project nodes, leading to inefficient project configuration that relies on human experience. Furthermore, the lack of data linkage between the R&D project management system and the financial system prevents automatic data collection and full-process traceability. Traditional project management tools are slow to respond to high-concurrency, large-scale project tasks, and the budget allocation results deviate significantly from the actual R&D situation.
By introducing named entity recognition technology based on a pre-trained bidirectional encoder representation model, the automatic conversion of policy text into structured node parameters is realized. A mathematical model of multi-dimensional business constraints is constructed and combined with a multi-objective optimization genetic algorithm to break the limitations of traditional linear budget allocation. An event-driven incremental calculation mechanism and a project node dependency topology graph are used for dynamic adjustment.
It significantly improved the accuracy of R&D node configuration, ensuring precise alignment between policy requirements and project execution, increased the accuracy of budget allocation, reduced computational load, and enhanced the system's stability and throughput in large-scale enterprise R&D matrices.
Smart Images
Figure CN122415033A_ABST