There is lot of data on the internet which is not indexed by our conventional search engines. This web content is what we call as Hidden web or Deep web. Users have to fill various forms to access this hidden content. So, there is a need to make an interface which help us to automatically fill the forms to access the hidden data. This can be done by extracting attributes from html pages and compare those attributes with users query and resulting attributes would be used to fill those forms automatically. For best results, ontology, specifically domain ontology can be used for filling the forms of hidden web.