Top-k Query Based on Map Reduce

Information explosion brings too much trouble to human lives. In order to improve the query efficiency of big data, we focus on the research ofa parallel algorithm, top-k query based on MapReduce. Firstly, we introduce the classic top-k algorithms and the basic theory of parallel programming MapReduce Model. Secondly, we put forward to the improved top-k algorithm and describe two core modules: word frequency statistics module, which includes 4detailed steps, and sorting output module. At last, we do some experiments to compare their runtime, the results show the algorithm's efficiency and availability and provide a feasible way to deal with the information explosion.