利用地名语义实现Web地震事件空间信息提取

针对Web新闻中地震事件地点要素提取的需求,提出了利用地名语义技术实现的地震事件空间信息提取的解决方案。通过构建地名本体库来支持标准化地名构建,通过句法分析、中文分词和词性标注识别地名,通过地名的标准化从地名本体库获得地名语义信息提取地震事件地点信息。同时,利用语义库及其推理机制完成地名标定范围内的地震事件搜索,成功地获取了中国境内近年的各省地市县的地震事件的空间分布情况。原型系统验证了论文方法的可行性和有效性。

Spatial Information Extraction of Web Seismie Event Based on Geographic Names Semantic Technology

Aiming at the needs of geographic feature extrac- tion of web news semantic event, the solution of the spatial information extraction from seismic event news is proposed based on geographic names semantic technology. Basic strate- gies including the standardization of geographic names is supported through the building of geographic names ontology library. Geographic names are identified through syntactic analysis, Chinese word segmentation and part of speech tag- ging. Seismic events location information is extracted through the standardization of geographic names from the semantic ontology library of geographical names. Meanwhile, based on the semantic library and its reasoning mechanism, the spatial distribution of the seismic events is successfully obtained in the provinces, cities and counties in China in recent years. The prototype system is completed to verify the feasibility and effectiveness of the methods proposed by this paper.