This research book is carried to response to research problem titled ¿Study of Information Retrieval Performance Refinement in Deep Web Mining¿. To facilitate effective access to the deep web, whereas the crawl-and-index techniques often used in widespread search engines now a days have been quite fruitful for the surface web, such an retrieval model might not be appropriate for the deep web. At the outset, crawling will expected face the limit of coverage that seems inherent because of the unseen likely and dynamic nature of web databases. Furthermore, indexing of the crawled data can face the barrier of mechanical heterogeneity across the range of deep web information.