Citation

Bhattacharijee, Arpita and Yee, Ting Choo (2025) Data Driven Retail Recommendations: A Study on Multi-Aspect Features for Enhanced Decision Making. In: 10th International Conference on Big Data Analytics, ICBDA 2025, 13 March 2025 - 15 March 2025, Hybrid, Singapore.

Text Data Driven Retail Recommendations_ A Study on Multi-Aspect Features for Enhanced Decision Making.pdf - Published Version Restricted to Repository staff only Download (2MB)

Abstract

In today’s data-driven world, recommending the most suitable retail business for a specific location requires a holistic approach that integrates multi-faceted information, including geospatial, demographic, and consumer behavior data, to enhance decision-making and optimize market engagement. This study begins with the development of comprehensive and insightful analytical datasets that integrate features from various aspects of location data. We have applied several feature selection methods to identify the most influential features for model performance. To automate the recommendation process, We employed four machine learning models—XGBoost, LGBM, CatBoost, and GBClassifier—of which LGBM and XGBoost consistently achieved strong performance in terms of accuracy, precision, recall, and F1 score across all experimental datasets. This study also compares the impact of different feature categories on the model’s learning process, with features trained both independently and the integrated optimal features selected by the Boruta method. The integration of features significantly improved the model’s classification performance, as shown by scores of MCC, Cohen’s Kappa, and balanced accuracy. Overall, the proposed retail analytics framework holds potential in accurately classifying retail businesses, addressing an under-explored research area and introduces a compre

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Machine learning, recommendation
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	22 Dec 2025 03:19
Last Modified:	22 Dec 2025 03:19
URII:	http://shdl.mmu.edu.my/id/eprint/15090

Downloads

Edit (login required)