情报杂志2025，Vol.44Issue(6)：168-176,192,10.DOI:10.3969/j.issn.1002-1965.2025.06.021
电信网络诈骗警情要素抽取与模式挖掘研究
Research on Alarm Texts Element Extraction and Pattern Mining of Telecom Network Fraud
摘要
Abstract
[Research purpose]This study proposes a novel method for extracting and mining alarm texts elements to enhance the han-dling efficiency of 110 alarm texts related to telecom network fraud.The method solves the problem that the existing research does not fully mine the police situation data,and provides valuable intelligence support to public security organs.[Research method]A fine-tuned UIE method with a small sample size minimizes computational resource consumption and alleviates manual labeling efforts.A comprehensive knowledge graph of 14 telecom network fraud types is constructed,and subgraph traversal with the gSpan algorithm uncovers criminal pat-terns and regularities.[Research result/conclusion]The proposed method achieves a precision rate of 83.6％for alarm text element ex-traction,outperforming existing methods.The application of the knowledge graph and pattern mining techniques effectively identifies hid-den criminal patterns,enhancing the intelligent analysis and information utilization capabilities of public security organs.关键词
公安情报/电信网络诈骗/警情要素抽取/知识图谱/模式挖掘Key words
public security intelligence/telecom network fraud/alarm texts element extraction/knowledge graph/pattern mining引用本文复制引用
刘卓娴,石拓,胡啸峰..电信网络诈骗警情要素抽取与模式挖掘研究[J].情报杂志,2025,44(6):168-176,192,10.基金项目
北京市自然科学基金项目"基于多源数据的北京市电信网络诈骗被害分析及精准预防策略研究"(编号:9244025)研究成果. （编号:9244025）
