Selectivity estimation on streaming spatio-textual data using local correlations

In this paper, we investigate the selectivity estimation prob- lem for streaming spatio-textual data, which arises in many social network and geo-location applications. Specifically, given a set of continuously and rapidly arriving spatio- textual objects, each of which is described by a geo-location and a short text, we aim to accurately estimate the cardinal- ity of a spatial keyword query on objects seen so far, where a spatial keyword query consists of a search region and a set of query keywords. To the best of our knowledge, this is the first work to ad- dress this important problem. We first extend two existing techniques to solve this problem, and show their limitations. Inspired by two key observations on the "locality" of the correlations among query keywords, we propose a local cor- relation based method by utilizing an augmented adaptive space partition tree (A²SP-tree for short) to approximately learn a local Bayesian network on-the-fly for a given query and estimate its selectivity. A novel local boosting approach is presented to further enhance the learning accuracy of lo- cal Bayesian networks. Our comprehensive experiments on real-life datasets demonstrate the superior performance of the local correlation based algorithm in terms of estimation accuracy compared to other competitors.