What is Bucket-Based Histogram

In the context of Knowledge Discovery in Databases, data reduction is a pre-processing step delivering succinct yet meaningful data to sequent stages. If the target of mining are data streams, then it is crucial to suitably reduce them, since often analyses on such data require multiple scans. In this chapter, we propose a histogram-based approach to reducing sliding windows supporting approximate arbitrary (i.e., non biased) range-sum queries. The histogram is based on a hierarchical structure (as opposed to the flat structure of traditional ones) and it results suitable to directly support hierarchical queries, such as drill-down and roll-up operations. In particular, both sliding window shifting and quick query answering operations are logarithmic in the sliding window size. Experimental analysis shows the superiority of our method in terms of accuracy w.r.t. the state-of-the-art approaches in the context of histogram-based sliding window reduction techniques.