EFFICIENT RELATIONAL CONTENT BASED SEARCH USING SEMANTIC PATTERN EXTRACTION

Abstract

A large data is generated in different organizations which are in text format. In such data the structured information is get shadowed in unstructured data. Measuring the semantic similarity between content and query is an important component in various tasks on the web search. The relation extraction, community mining, document mining and automatic metadata extraction is the various components in the web. Despite its usefulness the relatively measuring semantic similarity between the queries remains a challenging task. An empirical method is used to estimate semantic similarity using query and text snippets retrieved from a web search engine is a relative task in document of information. To improve the search accuracy various word co-occurrences is measured and integrate those data with the lexical pattern where the text is extracted.