MARKET 4.0: EXPLORING MARKET BASKET ANALYSIS (MBA) ALGORITHMS FOR CO-MARKET INTELLIGENT APPLICATION
Daniel T. Ursulum, Jr.,
Thelma D. Palaoag
Abstract: This study was initiated with thorough investigation to determine the optimal Market Basket Analysis (MBA) algorithms for the Co-Market Intelligent Application using Association Rule Mining (ARM). An exploration evaluation was conducted to assess their suitability for integration into the Co-Market Intelligent application framework, focusing on three product categories: coconut non-food products, coconut food products, and sarakat products. Employing a model for Apriori with a minimum support of 0.2 and minimum confidence of 0.5 yielded remarkably high accuracies of 140% for coconut non-food and 72% for sarakat products while a model with minimum support of 0.2 and minimum confidence of 0.7 yielded 92% lift ratio for food products. For FP-Growth, a model with minimum support of 0.1 and minimum confidence of 0.5 demonstrated the best performance, achieving accuracies of 92% and 42% for coconut non-food and food products respectively. The findings of this study suggest that the selection of either the Apriori or FP-Growth algorithm, or a combination of both, tailored to the specific product categories, can significantly enhance the efficiency and accuracy of Market Basket Analysis in the Co-Market Intelligent Application, providing valuable insights for optimized decision-making and strategic planning in the targeted market sectors.
Keywords: Intelligent Application, MBA (Market Basket Analysis), Apriori Algorithm, FP-Growth Algorithm, Association Rule Mining
DOI: 10.24874/PES06.04.011
Recieved: 12.03.2024 Revised: 09.07.2024 Accepted: 31.07.2024
UDC: 004.021:339.13
Reads: 935 
