Quality based Web information extraction approach using NLP and Text Mining

Abstract: Today for discovering information regarding any matter people give preference to web resources rather than other resource. But World Wide Web contains billions of web pages and various other documents containing information about various topics. So we use a search tool for retrieving information about any particular topic. And performance of these search tools is mostly described by the quality of search result. For improving quality of search result search strategies uses various techniques. Despite of these techniques the quality of search results is degraded due to uncertainty in intending context of fired query. Another problem of search result is redundant pages that contain contents that are already contained by previous pages; degrade the quality of search result. In this paper we proposed an approach for improving quality of search results with the help of Natural language Processing and Text Mining. Here quality of search results is described by relevancy precision of search results return by search strategy. The relevancy precision can be defined by ratio of relevant articles to total retrieved articles.

Keywords: WWW, NLP, WSD, SYNSETs, SERPs