Adaptive Technique for Document Annotation to Identify Attributes of Interest

Abstract

Many application domains generate and share information which describes their products and services. Such description contains unstructured information. So, it is always difficult to find the useful metadata. The information extraction algorithms are very expensive or inaccurate when operating on such unstructured information. This paper proposes adaptive technique for document annotation process to retrieve the useful information. This approach is based on Collaborative Adaptive Data Sharing (CADS) platform for document annotation. A CADS uses query workload to direct the annotation process. A key attribute of CADS is that it identifies important data attributes of the application. Further it uses this information to direct the data insertion and querying.