Now a days so many organization create and share the textual description of their products or service and action etc. it is contains for most amount collection of structured data and which is remains worried about unstructured the information, if data extraction structural relation by using algorithms facilitating, they are more cost and inaccurate information. When is working top of text, it does not is contains structural information. An anther approach to the generating of the structure of metadata by the identifying that documents, that is likely to contain information of interest. That data are going to be valuable for questioning information based used. These approaches based on the idea that humans are more likely to add the necessary metadata during generate the time. This process based on the collaborative adaptive data sharing platform[CADS] approach to query workload by up to 50 percent only visibility of document. So further probing algorithm with Bayesian approach technique was included, that can be improve the efficient of visibility of document or data with respect the query and content workload based on the more than 50 percent improve.