Abstract

There are many organizations which generate and share textual description of their product, this description contain large amount of structured data which remains buried under unstructured data. There are many extraction algorithms which can extract the structured data but this algorithms are expensive and inaccurate. So we present an alternative which enables the generation of structured metadata by identifying the documents which contain the useful information and this information can be used for querying the data base ie annotation of document .As there are thousands of attribute some may have the common meaning but different names . This limitation makes analysis and querying of database based on annotation cumbersome Based on these annotations we can give the queries to extract the required document. This approach will help us in proper analysis and searching of the document, in an accurate way.

Paper Details