Optimizing Retail Recommendation via Similarity Measures and Machine Learning Approach

Citation

Bhattacharijee, Arpita and Ting, Choo Yee and Ghauth, Khairil Imran and Peng, Loh Yuen and Hashim, Noramiza and Mohd Isa, Wan Noorshahida and Suvon, Injamul Haque and Matsah, Wan Razali (2024) Optimizing Retail Recommendation via Similarity Measures and Machine Learning Approach. JOIV : International Journal on Informatics Visualization, 8 (3). pp. 1192-1202. ISSN 2549-9610

[img] Text
2360-8153-1-PB.pdf - Published Version
Restricted to Repository staff only

Download (4MB)

Abstract

Finding a suitable retail business with potential success in a specific location can be challenging for retailers. The process is often lengthy and inconsistent due to the subjective nature of expert opinions. Previous research has demonstrated several techniques that consider numerous influential attributes for location optimization problems. However, while many studies rely on a business's core data for analytical purposes, accessing this information is often a significant constraint. This study aims to address the challenge of extracting valuable location features to enhance the profitability of chosen businesses despite the inaccessibility of core business data. The proposed methodology involves three main steps. First, an analytical dataset must be created by utilizing geographic and demographic information. Second, we conduct similarity measures by applying Manhattan distance to the entire analytical dataset, using an ideal business outlet that contains the footfall information. Through this process, we can identify businesses that share similar characteristics with popular outlets. Finally, several supervised machine learning models are trained to employ the extracted meta-features. Experimental results show that the XGBoost classifier performs best with an 87% accuracy score, outperforming the baseline models. The proposed methodology in this research presents a robust framework that demonstrated remarkable efficiency in achieving the stated objectives and improving the performance of retail business recommendations within a given location. Future work could consider a broader range of features that could potentially enhance model performance by applying ensemble learning.

Item Type: Article
Uncontrolled Keywords: location optimization; analytical dataset; footfall information; similarity measure; machine learning; recommendation
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 04 Nov 2024 02:32
Last Modified: 04 Nov 2024 02:32
URII: http://shdl.mmu.edu.my/id/eprint/13119

Downloads

Downloads per month over past year

View ItemEdit (login required)