A method and system for analyzing transformation of scientific and technological achievements based on reinforcement learning

By constructing a reinforcement learning model and preprocessing multi-source data, the problem of insufficient processing of multi-source data was solved, realizing intelligent and dynamic decision-making in the analysis of scientific and technological achievements transformation, and improving the executability and stability of the analysis results.

CN122134504APending Publication Date: 2026-06-02WUXI SHENSHAN XINTIAN TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
WUXI SHENSHAN XINTIAN TECHNOLOGY CO LTD
Filing Date
2026-02-06
Publication Date
2026-06-02

Smart Images

  • Figure CN122134504A_ABST
    Figure CN122134504A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for analyzing the transformation of scientific and technological achievements based on reinforcement learning, comprising the following steps: acquiring and preprocessing multi-source data associated with the target scientific and technological achievement; constructing the state space of the transformation process; constructing the action set of the transformation process; generating the reward value of the transformation process; constructing and training an improved AIRL model to obtain the trained improved AIRL model; outputting the corresponding analysis results of the transformation of scientific and technological achievements; and updating the trained improved AIRL model. This realizes dynamic modeling and intelligent analysis of the transformation process of scientific and technological achievements, improving the scientificity, adaptability, and feasibility of the transformation decision-making.
Need to check novelty before this filing date? Find Prior Art