Selectivity Estimation on Set Containment Search

Abstract

In this paper, we study the problem of selectivity estimation on set containment search. Given a query record Q and a record dataset S, we aim to accurately and efficiently estimate the selectivity of set containment search of query Q over S. The problem has many important applications in commercial fields and scientific studies. To the best of our knowledge, this is the first work to study this important problem. We first extend existing distinct value estimating techniques to solve this problem and develop an inverted list and G-KMV sketch based approach IL-GKMV. We analyse that the performance of IL-GKMV degrades with the increase of vocabulary size. Motivated by limitations of existing techniques and the inherent challenges of the problem, we resort to developing effective and efficient sampling approaches and propose an ordered trie structure based sampling approach named OT-Sampling. OT-Sampling partitions records based on element frequency and occurrence patterns and is significantly more accurate compared with simple random sampling method and IL-GKMV. To further enhance performance, a divide-and-conquer based sampling approach, DC-Sampling, is presented with an inclusion/exclusion prefix to explore the pruning opportunities. We theoretically analyse the proposed techniques regarding various accuracy estimators. Our comprehensive experiments on 6 real datasets verify the effectiveness and efficiency of our proposed techniques.

Ext-LOUDS: A Space Efficient Extended LOUDS Index for Superset Query

“…Efficient indexes are key to improving superset query performance. Inverted index [4,5] is one of the most common indexes used in this field. Helmer et al [6] designed an inverted index which scans all relevant inverted lists…”