基于深度学习的档案信息化整合方法及系统
By using deep learning technology to preprocess and semantically analyze archival data and construct a knowledge graph, the problem of unified management and secure utilization of archival management systems under heterogeneous and multimodal data is solved. This enables efficient semantic querying and refined access control, thereby enhancing the utilization value of archival information.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA SHENHUA ENERGY CO LTD SHENDONG COAL BRANCH
- Filing Date
- 2025-12-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing record management systems struggle to achieve unified management and efficient utilization when faced with massive, heterogeneous, and multimodal data. They have limited semantic understanding capabilities, failing to meet users' in-depth and multi-dimensional query needs, and are inadequate in terms of data security and access control.
A deep learning-based approach to archival information integration is adopted, including modules for archival preprocessing, semantic index construction, knowledge graph construction, semantic relationship retrieval, and access control. Through format parsing, text normalization, image correction, and metadata verification, a standardized archival preprocessing dataset is generated, and an archival semantic vector set and knowledge graph are constructed to support cross-modal queries and refined access control.
It enhances the semantic understanding and retrieval efficiency of archival content, supports complex semantic queries, realizes advanced semantic retrieval and knowledge discovery, meets diverse user needs, and maximizes the value of archival information while ensuring data security.
Smart Images
Figure CN121706140B_ABST