Dynamic query forms (DQF) using ranking models for database queries

摘要

The widespread use of databases in government, industry, and education, and the development of the Internet and World Wide Web result into large and complex databases. Modern science and scientific database also have lots of attributes. The static and automatic query forms are not able to get user required result on these huge databases. This paper proposes a query form interface, which is able to dynamically generate query forms called as Dynamic Query Forms (DQF) with the ranking model to enhance the feature with ranked form components. The DQF captures feedback and reference from the user ad those are used to rank query form components which assist the user in making decisions of choosing the highest ranked component. In this paper, we used three different ranking models 1) Probabilistic ranking model, 2) learning to rank model by using frequent pattern tree and 3) Bayesian network model. The iterative process is used to generate a query form and is guided by the user. The system automatically generates ranking lists of form components at each of the iteration by using ranking model chosen by the user. And the user can update or enhance the query form using those ranked query form component to get the desired result. After enhanced updated query form the user can submit those queries to view the result at each of the iteration. Our experimental evaluation compares between three ranking models and demonstrates effectiveness and efficiency of the system.