摘要
为了解决从网页中准确抽取产品信息这一B2B垂直搜索引擎的关键问题,以站点树为模型,首先分析了企业网站的结构特征,在此基础上构建了一个面向B2B垂直搜索引擎的网页信息抽取系统。该系统利用站点树在企业站点大量网页中识别出产品页,并进行去噪处理,然后使用基于规则的方法抽取产品页中包含的产品描述信息和参数信息。通过该系统抽取到的各类产品信息较为准确,且效率得到明显提高,适用于B2B垂直搜索引擎中对产品的描述、分类及搜索。
To solve the problem of information extraction on web pages, which is one of the key technologies of B2B vertical search en- gine,taking website as model, structure of the corporation website is analyzed firstly,based on which a system of web information extrac- tion for B2B vertical search engine is constructed. The website tree is used in the system for identification and noise elimination of the product pages, and then description and parameter information of the products contained in product pages are extracted according to the rules. All kinds of information extracted accurately and efficiently by the system can be used for description, classification and searching of the products in B2B vertical search engine.
出处
《计算机技术与发展》
2013年第2期153-156,161,共5页
Computer Technology and Development
基金
中央高校基本科研业务费专项基金项目(12zy019)
关键词
B2B垂直搜索
网页信息抽取
企业站点树
去噪
B2B vertical search engine
web information extraction
corporation website tree
noise elimination
