Paper Contents
Abstract
Traditional e-commerce advertising often lacks personalization, leading to wasted marketing spend and a poor user experience from irrelevant ads. This project, "AdOptima: Predictive Ad Engagement System for E-Commerce Platforms," addresses this by creating a system for targeted advertising within the E-Shopper platform. Leveraging machine learning on user data, AdOptima predicts whether a user is likely to click an ad (using a Decision Tree model) and, if so, their most probable product category of interest (using a Random Forest model). This predictive intelligence enables dynamic ad display decisions based on individual user potential and predicted interest.The system is implemented as a web-based application using Python with Flask for the backend and MySQL for data storage. It integrates with the E-Shopper frontend to show targeted ads on key pages based on the predictions and includes an Admin Panel for monitoring performance and user data. The project evaluates model accuracy using standard metrics and aims to validate the business impact of targeted advertising through quantifiable improvements in key metrics like Click-Through Rate.
Copyright
Copyright © 2025 MEGANATHAN M. This is an open access article distributed under the Creative Commons Attribution License.
