An overview of data structures and algorithms: case study of us in the vector-space model and mining off requentitem sets using the apriori algorithm

Abstract

In this paper, we review some commonly used data structures and algorithms. We then review two important problems: the creation of the vector-space model that is widely used in the design of information retrieval systems, and the mining of frequent itemsets using the apriori algorithm. We consider two variations of the apriori algorithm: the first is the classical algorithm which computes candidate k-itemsets by first joining frequent (k-1)-itemsets to themselves, and applying the apriori property to prune the generated candidate k-itemsets; the second avoids the join stage in the classical algorithm, and instead, generates candidate k-itemsets directly from rows of the transactions database, followed by application of the apriori property to prune each itemset so determined. Finally, we illustrate appropriate data structures and algorithms that when put together, provide efficient implementations of our solution to the problems mentioned.

Keywords: data structures, algorithms, vector-space model, frequent itemsets mining, apriori algorithm