人才网页自动识别系统研究

Study on Talents Description Web Page Automatic Recognition System

提出人才网页自动识别系统设计,实现对Nutch定向采集系统抓取的高校网站页面进行人才描述网页自动识别。识别过程中使用自动获取的网页的URL特征、网页Title标签特征、链接文字特征以及网页文本内容特征,使用人名词表、正面特征词表、负面特征词表对各项识别特征进行匹配以计算特征值,借助开源软件LibSVM实现基于多特征值的人才网页自动识别。

The paper brings forward a talents description Web page automatic recognition system, realizes automatic recognition methods of university talents description Web pages which are crawled by Nutch crawl system. During the automatic recognition process, features of Web page URL, title label content, anchor text content and Web page content are used. The value of those features are computed based on matching of name list, positive feature word list and negative feature word list. Based on multiple feature values, the system uses LibSVM to realize talents description Web page automatic recognition.