|
Performance comparison of association rule mining algorithms among Apriori, FP-Growth, FP-Max, and H-Mine for market basket analysis |
|---|---|
| รหัสดีโอไอ | |
| Creator | Kritbodin Phiwhorm |
| Title | Performance comparison of association rule mining algorithms among Apriori, FP-Growth, FP-Max, and H-Mine for market basket analysis |
| Publisher | Mahasarakham University |
| Publication Year | 2569 |
| Journal Title | Journal of Science and Technology Mahasarakham University |
| Journal Vol. | 45 |
| Journal No. | 1 |
| Page no. | 53-61 |
| Keyword | Association rule mining, Apriori, FP-Growth, FP-Max, H-Mine |
| URL Website | https://li01.tci-thaijo.org/index.php/scimsujournal |
| Website title | Journal of Science and Technology Mahasarakham University |
| ISSN | 1686-9664 (Print), 2586-9795(Online) |
| Abstract | Association rule mining is a crucial technique for market basket analysis in retail businesses, but it often faces challenges in processing speed and memory usage, particularly with large-scale datasets. This research presents a performance comparison of four algorithms: Apriori, FP-Growth, FP-Max, and H-Mine, using a grocery store dataset for market basket analysis under varying support thresholds. The results showed that the H-Mine algorithm demonstrated superior performance in both execution time and memory usage, attributed to its efficient Hyperlink data structure, followed by FP-Growth and FP-Max algorithms, which employ FP-Tree structure to minimize database scanning. Meanwhile, the Apriori algorithm exhibited the lowest performance. |
