Template Based Relevant Information Extraction from Web Pages

Abstract

Extracting relevant data at web scale have variety of application which include useful data extraction, improve the quality of web search, collection of template, information collection and comparing data. With the growth of the Internet, there has been a rapid growth of online resources and information. But the information on internet is not in relevant format. Data on web are in unstructured form, so there should be some system which can extract relevant structured data from this instructed web data. There are various techniques to extract information at web scale using template extraction. TEXT template extraction technique is used for extraction and detection of template, but it extracts the entire site and work on static web pages. We propose a template - based information extraction approach to address the issues mentioned above. Our Information extraction system has algorithms for web page clustering, detecting site change, rule relearning, it is the system to do high accuracy information mining at web scale.