Assisting document annotation using a data sharing platform

Most of the organizations today generate and share textual descriptions of their products, services, and actions. Such textual collections consists of significant amount of structured information, which remains buried in the unstructured text. Even though information extraction algorithms provides the extraction of structured relations, they are often expensive and inaccurate. We present an alternative approach that facilitates the generation of the structured metadata which is useful for querying the database. Document annotations are comments, notes, or other types of external remarks that can be attached to a document. Our approach relies on the idea that humans can add the valuable metadata during creation time prompted by an interface. We can increase the quality of searching to some extent, if the documents are properly annotated. Existing probabilistic approaches have some demerits. This paper proposes an approach that uses inverted indexing scheme for effective mining of informations. Major contribution of this paper relies on the fact that annotates the document by utilizing the content of the document and query workload.