PERBANDINGAN KINERJA ALGORITMA APRIORI DAN ECLAT (Equivalence Class Transformation) DALAM MENERAPKAN REKOMENDASI BARANG DISKON PADA DATA TRANSAKSI

Pamungkas, Elang Damar Galih (2025) PERBANDINGAN KINERJA ALGORITMA APRIORI DAN ECLAT (Equivalence Class Transformation) DALAM MENERAPKAN REKOMENDASI BARANG DISKON PADA DATA TRANSAKSI. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) ppppppppppppppppppppppppppppppppppppppp-1-21.pdf Download (715kB)
	Text (Bab 1) Bab 1.pdf Download (96kB)
	Text (Bab 2) Bab 2.pdf Restricted to Repository staff only until 26 May 2027. Download (619kB)
	Text (Bab 3) Bab 3.pdf Restricted to Repository staff only until 26 May 2027. Download (917kB)
	Text (Bab 4) Bab 4.pdf Restricted to Repository staff only until 26 May 2027. Download (1MB)
	Text (Bab 5) Bab 5.pdf Download (85kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf Download (136kB)

Abstract

This study aims to compare the performance of the Apriori and ECLAT algorithms in generating discount item recommendations based on sales transaction data from CV SOSO Group Jombang. The research is conducted in six systematic stages: collecting transaction data from October, preprocessing the data into basket format, implementing the Apriori and ECLAT algorithms, generating association rules using support, confidence, and lift metrics, evaluating algorithm performance, and visualizing the results. Performance evaluation compares the number of itemsets generated, execution time, and the quality of the resulting rules. Experiments are conducted using three minimum support values (0.1, 0.03, and 0.05) across datasets of various sizes. The results show that Apriori performs better on small to medium-sized datasets, with faster execution times and more interpretable rules for decision-makers. Conversely, ECLAT demonstrates superior computational efficiency and memory usage on larger datasets and is more effective at uncovering strong product associations with high support. In terms of rule quantity, Apriori tends to generate more itemsets at higher support thresholds, while both algorithms yield similar outcomes at lower thresholds. In conclusion, algorithm selection should align with specific objectives: Apriori is more suitable for marketing strategies requiring interpretable rules, while ECLAT is recommended for efficient pattern discovery in large-scale transaction data. Keywords: Apriori, ECLAT, Market Basket Analysis, Discount Recommendation, Data Mining.

Item Type:

Thesis (Undergraduate)

Contributors: