一种基于SVM的Web信息自动化抽取方法

Web Information Automatic Extraction Method Based on SVM

针对传统的Web信息抽取方法运算量大、自动化程度低的问题,提出了一种基于SVM的WEB信息自动化抽取方法。利用SVM优秀的分类性能将网页中有用数据和无用数据分类标注,有效地完成Web信息抽取任务,准确地抽取出所需信息,实现数据抽取的自动化。实验结果表明,该方法可以有效地获取网页信息特征,具有较高的召回率和准确率。

Given the problems of heavy computation and low automatic level existed in traditional Web information extraction method, this paper proposes a kind of web information automatic extraction method based on SVM, effectively completing the task of Web information extraction and precisely extracting information so as to realize automation of data extraction. The results show that SVM can be used in web information extraction and it has higher rates of recall and precision.