Publication Date

Fall 2025

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Engineering

Advisor

Bernardo Flores; Jun Liu; Magdalini Eirinaki

Abstract

Apriori is a machine learning algorithm developed in 1994 by R. Agrawal and R. Srikant for association rule mining purposes. This family of algorithms takes transactional data and analyzes relationships between variables in large datasets. The typical output of such algorithms is a prediction that if users choose item X, it is highly likely that they will also choose item Y. Apriori is known to be a robust algorithm and is used by many large companies in order to analyze user tendencies and even make recommendations. Although Apriori is a powerful algorithm, its original implementation is known to have limitations, especially as the volume of data grows. Due to multiple database scans, the resources are being used inefficiently, and the overall performance becomes a big problem. Even though there are certain hyperparameters that help regulate Apriori, the list of candidate items generated in between the input and output tends to be unjustifiably large, often wasting memory. Last but not least, the output itself is often too big and would take up unacceptable amounts of storage space. Due to all these issues but great performance nonetheless, a lot of engineers opt to use modified versions of Apriori. The purpose of this thesis is to assess the original Apriori algorithm, as well as its most popular modifications, and propose a new algorithm. The new algorithm aims to address existing issues of Apriori and improve its performance, which will provide for faster and more efficient association rule mining.

Download

Share

COinS