Exact query on big data is a challenging task due to the large numbers of autonomous data sources. In this paper, an efficient method is proposed to select sources on… Click to show full abstract
Exact query on big data is a challenging task due to the large numbers of autonomous data sources. In this paper, an efficient method is proposed to select sources on big data for approximate query. A gain model is presented for source selection by considering information coverage and quality provided by sources. Under this model, the source selection problem is formalized into two optimization problems. Because of the NP-hardness of proposed problems, two approximate algorithms are devised to solve them respectively, and their approximate ratios and complexities are analyzed. To further improve efficiency, a randomized method is developed for gain estimation. Based on it, the time complexities of improved algorithms are sub-linear in the number of data item. Experimental results show high efficiency and scalability of proposed algorithms.
               
Click one of the above tabs to view related content.