Theses and Dissertations Collection

Throughout the years, researchers have striven to improve the quality of information retrieval from the web; especially for ordinary users who need the proper access to the desired information. Users mostly interact with the web through search engines and they tend to expect precisely what they asked for. As web databases (Deep Web) hold enormous amounts of high quality information that users need to access and leverage, we shed some light on the importance of searching the deep web rather than just querying it. Searching is more flexible than querying based on fixed variables. Searching the deep web can enhance information accessibility especially if it simulates user behaviour. Research interest is growing in the area of maximizing the usefulness of web search via utilizing the largest portion of the web (Deep Web). Search engines such as Google can only find indexed information that is present in the Shallow Web. In contrast, peeking into Deep Web databases is not possible for search engines such as Google. Search engines are not able to simulate SQL-like queries of database contents in a traditional or other intuitive human-like manner. In this work, we present a system called DeepQ to search the database contents behind the firewalls inside the deep web and show that these contents can be accessed using a structured query language treating them as a deep relational web. We leverage a recent proposed declarative deep web query language called DQL, and we present the contours of its implementation in the DeepQ system. We also believe that this work has the potential to demonstrate the Universal Relations model, as the user will be able to interact with databases that are hidden behind firewalls, and also search conveniently for information.