基于本体的Web信息抽取系统

摘要

为了解决已有信息抽取系统中方法不具有重用性及不能抽取语义信息的问题,提出了一个基于领域本体的面向主题的Web信息抽取框架.对Web中文页面,借助外部资料,利用本体解析信息,对文件采集及预处理中的源文档及信息采集、文档预处理、文档存储等技术进行了分析设计,提出了文本转换中的分词及词表查询和命名实体识别算法,并给出了一种知识抽取方案.实验结果表明,该方法可以得到性能较高的抽取结果.%To address the semantic problem and method reuse of traditional information extraction system, a topic-oriented Web information extraction framework based on domain ontology is proposed. For Chinese documents on the Web, with external data and domain ontology, the source document, information collection, document pre-processing, document storage and document database of the document collection and preprocessing are analyzed, the word segmentation, vocabulary queries and named entity recognition algorithms of text conversion is introduced, In the end, a knowledge extraction method is given. The test shows that the method can get higher performance results.