Method and device for extracting and clustering information of appeal cases based on public opinion big data
By constructing an entity information extraction model based on ERNIE pre-trained language models and pointer networks, and combining representational and interactive matching networks, the problem of secure extraction and clustering of complaint case information in public opinion big data was solved, achieving fast and accurate information processing and privacy protection.
Patent Information
- Application Number
- CN202211503457.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-28
- Publication Date
- 2026-04-14
- Estimated Expiration
- 2042-11-28
AI Technical Summary
Existing technologies struggle to effectively and securely extract and cluster entity information from public opinion big data, especially while protecting user privacy. Furthermore, traditional methods are inefficient and inaccurate with massive amounts of data.
We construct an entity information extraction model based on a pre-trained language model and pointer network of ERNIE, combine it with representational and interactive matching networks, and use the DBSCAN algorithm to match and cluster case entity information through deep learning and unsupervised clustering methods, avoiding the use of third-party geocoding services to protect privacy.
It enables the rapid and accurate extraction and clustering of key information from large-scale public opinion big data, improving processing efficiency and protecting user privacy.