Processing Recommender Top-N Queries In Relational Databases

With the development of Internet technology, many people who are hold that mainlythrough the interaction of search engine to get the information, the traditional search engines,such as Google, Yahoo, Baidu and so on, the user mainly through input the contents of thequery in the search engine, the query system according to the query keywords to retrieve theresults from the relational database and returns to the user. The relational database keywordsquery not need to know the relational database model and complex query language, such asSQL statement. Relational database keywords query mainly based on two kinds of methods,based on the data graph and based on the model graph. These two methods are how to find thesmallest candidate tuples result tree. In search engine, the query recommended isreestablishing new query, in order to meet the needs of users. A good recommendation queryis not only related to the initial query, but also combined with the user’s query intention.Query recommendation is one of the essential technologies in the search engine, and inorder to find out the other keywords associated with the original query. According to thefeedback information by a user in the result sets of initial/previous queries, in this paper wepresent a framework for processing recommender top-N queries in relational databases. Basedon the techniques and ranking strategies of keyword search, this framework returns top-Nresults for an initial query given by the user. As soon as he/she selects some of the top-Nresults, the framework will find out related keywords from the results selected by the user,calculate and modify corresponding weights of the related keywords. By using the weights,our framework determines new query words associated with the initial query to construct arecommender query. A knowledge base is created to store the related information of the tuplesin the underlying database for evaluating the recommender query. The experimental resultsbased on real datasets show the efficiency and effectiveness of our framework.