
Research Article
Predicting Probable Product Swaps in Customer Behaviour: An In-depth Analysis of Forecasting Techniques, Factors Influencing Decisions, and Implications for Business Strategies
@ARTICLE{10.4108/eetsis.4049, author={Mohit M Rao and Virendra Kumar` Shrivastava}, title={Predicting Probable Product Swaps in Customer Behaviour: An In-depth Analysis of Forecasting Techniques, Factors Influencing Decisions, and Implications for Business Strategies}, journal={EAI Endorsed Transactions on Scalable Information Systems}, volume={10}, number={6}, publisher={EAI}, journal_a={SIS}, year={2023}, month={10}, keywords={Prediction, Product swap, Feature Selection, Random Forest, ranking, chi-square test, Support Vector Machine, Machine Learning, Artificial Intelligence}, doi={10.4108/eetsis.4049} }
- Mohit M Rao
Virendra Kumar` Shrivastava
Year: 2023
Predicting Probable Product Swaps in Customer Behaviour: An In-depth Analysis of Forecasting Techniques, Factors Influencing Decisions, and Implications for Business Strategies
SIS
EAI
DOI: 10.4108/eetsis.4049
Abstract
Introduction: Factors influencing product swap requests and predict the likelihood of such requests, focusing on product usage, attributes, and customer behaviour, particularly in the IT industry. Objectives: Analyse customer and product data from a leading IT company, aiming to uncover insights and determinants of swap requests Methods: Gather product and customer data, perform data processing, and employ machine learning methods such as Random Forest, Support Vector Machine, and Naive Bayes to discern the variables influencing product swap requests and apply them for classification purposes. Results: Analysed a substantial dataset, comprising 320K product purchase requests and 30K swap requests from a prominent social media company. The dataset encompasses 520 attributes, encompassing customer and product details, usage data, purchase history, and chatter comments related to swap requests. The study compared Random Forest, Support Vector Machine, and Naïve Bayes models, with Random Forest fine-tuned for optimal results and feature importance identified based on F1 scores to understand attribute relevance in swap requests. Conclusion: Evaluated three algorithms: support vector machine, naive Bayes, and Random Forest. The Random Forest, fine-tuned based on feature importance, yielded the best results with an accuracy of 0.83 and an F1 score of 0.86.
Copyright © 2023 M. M. Rao et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.