AI-Powered Data Analytics for Strengthening Customer Loyalty and Personalization in Retail

Abstract

An essential prerequisite for the growth of client loyalty is the assessment of customer loyalty. Over the past few decades, marketing academics have created a variety of client loyalty assessment tools. Because service marketplaces are getting more competitive, businesses are having trouble bringing in and keeping clients. Consequently, there is an increasing need to encourage client loyalty. This study presents a machine learning (ML) approach to customer loyalty prediction using the Online Retail II UCI dataset, a sizable collection of actual transactions. The data set was preprocessed by addressing missing data, data outliers, and data normalization using Z-score to enhance data quality. The Multilayer Perceptron (MLP) model has been created and assessed based on the indices like R 2, RMSE, and MSE indicating the good predictive ability of the model with an R 2 value of 93.98. Additionally, the comparison research showed that the MLP outperformed other models. The results point to the efficiency of advanced methods of learning in cognizing customer behavior and data-driven decision-making in the retail field.