Hash join algorithms used in text-based information retrieval: guidelines for users

Abstract

Text-Based Information Retrieval (IR) is a field where the search in a large document is a basic concept. Concise queries are very fundamentals process in order to satisfy user's need for information. One of the basic fundamental techniques in IR to implement queries is Hashing. Different types of hashing algorithms are used in IR. This paper discussed about guidelines for users who are implementing Hash join algorithm variations in their IR applications. Algorithms are varied based on its techniques involved in join operations. Three different variations of hash join algorithm, namely, XJoin algorithm, Hash Merge Join (HMJ) algorithm, and Early Hash Join (EHJ) algorithm are studied experimentally. Analysis on the results obtained is given. A user guideline based on three factors: overall execution time, response time and input/output operations performed are presented.