Skip to content

Research Overview

Course: COS-781 (Data Mining)
Research Topic: Improved Apriori Algorithm and FP-Growth Algorithm
Focus: Efficient frequent itemset mining

This research project aims to:

  • Analyze the limitations of the traditional Apriori algorithm
  • Study the FP-Growth algorithm and its advantages
  • Propose and implement improvements to enhance Apriori performance
  • Compare improved Apriori with FP-Growth algorithm
  • Evaluate both algorithms against standard benchmarks
  • Document findings and present results

The Apriori algorithm is a fundamental algorithm in association rule mining, but it has known limitations in terms of efficiency, especially with large datasets. FP-Growth provides an alternative approach that avoids candidate generation. This research explores methods to improve Apriori’s performance while maintaining correctness, and compares it with FP-Growth to understand the trade-offs between different approaches.

  • A comprehensive analysis of the Apriori and FP-Growth algorithms
  • An improved version of Apriori with documented enhancements
  • Comparative analysis between improved Apriori and FP-Growth
  • Experimental results comparing performance
  • Documentation and presentation materials