A New Method to Find Top K Items in Data Streams at Arbitrary Time Granularities

Abstract

Finding top K items in data streams means finding K items whose frequence are larger than other items in data streams. There are some methods to find most frequent K items in the whole data streams, but they can't be used in arbitrary time interval. This paper proposes a new method-MMF(K)_MS to find most frequent K items based on Hierarchical Synopsis. MMF(K)_MS supports query in arbitrary time interval through using HFVN framework with variable number of node in every layer and using Count Stretch data structure to maintain Synopsis in each layer. At Last, Proving MMF(K)_MS rational and available by experiment.