Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{201131,
        author = {Devilavanya S and Sri Bhagavathi M and Roshan Akthar M R and Elakya G},
        title = {Personalized E-Commerce Product Recommendations with Counterfactuals},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {4322-4331},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201131},
        abstract = {The use of artificial intelligence on e-commerce websites like Amazon and Flipkart to provide personalized product recommendations is very common. Although such systems are very predictive, they are usually not very transparent and the users do not know why they suggest certain products. Lack of explainability and lowers user trust and restricts user understanding of how preference or behavior change may affect recommendations. Precision and recall are the most common performance metrics that most traditional recommendation systems focus on, not taking into account the relevance of interpretability and user-centered insights. 
The proposed paper is a new framework named Personalized E-Commerce Product Recommendations with Counterfactuals that incorporates the use of counterfactual explainable AI into the recommendation engines. The system does not just predict the products that are relevant but also produces directions that are intuitive in the form of what-if explanations that display how small changes in the user behavior, preferences or filters would change the recommended results. This method maximizes transparency by introducing low and significant changes to accomplish various recommendations and gives the user a chance to take action based on the findings. 
The suggested model is a hybrid of collaborative filtering and content-based algorithms, a counterfactual explanation module, to make it accurate and interpretable. Experimental analysis of the findings proves that the integration of counterfactual explanations enhances user trust, satisfaction, and engagement with little harm to the performance of the recommendations. This work is part of the creation of transparent, trusting, and user-friendly AI solutions to the world of e-commerce.},
        keywords = {E-Commerce, Personalized Recommendation System, Explainable AI, Counterfactual Explanation, User Trust, Collaborative Filtering, Transparent AI, Interpretable Machine Learning.},
        month = {May},
        }