Leveraging Machine Learning Techniques to Project Customer Behaviour Through Predictive Analysis and Ethical Marketing

Abstract

Predictive analytics stands as a cornerstone of customer-centric strategies, offering invaluable insights. While customer analysis has been conducted for years, the manual handling of data has limited its effectiveness. Using predictive analytics tools, marketers have the potential to manipulate customers unethically, grooming them to purchase products they wouldn't otherwise consider buying. This research investigates the intricate dynamics of consumer behaviour and the transformative impact of predictive machine learning algorithms. Employing a mixed-methods research design combining quantitative and qualitative techniques, the study explores the application of unsupervised K-means clustering and supervised random forest algorithms. Through real-world case studies and data analysis, insights are gained into the predictive modelling of customer behaviour in diverse industries. Findings reveal the effectiveness of these techniques in segmenting customers based on income and spending behaviour, with a prediction accuracy of 84%. Furthermore, the study underscores the importance of integrating qualitative insights to enrich understanding and validity. The study also critically explores the potential risks associated with unethical marketing that led customers to purchase products without their voluntary and fully informed consent. This research contributes to advancing the understanding of consumer behaviour forecasting and predictive machine learning applications, paving the way for future research endeavours in this domain.

Citation

Shah, M.A. and Kumar, P. (2024), "Leveraging Machine Learning Techniques to Project Customer Behaviour Through Predictive Analysis and Ethical Marketing", Dadwal, S.S., Jahankhani, H. and Revett, K. (Ed.) Market Grooming, Emerald Publishing Limited, Leeds, pp. 121-138. https://doi.org/10.1108/978-1-83549-001-320241006