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{174147,
        author = {Chinmayee Bondgulwar and Sakshi Jagtap and Sarthak Shelar and Pratiksha Shevatekar},
        title = {Eco-Friendly Product Recommendation System Using  Large Language Model Lllama-2},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {10},
        pages = {2699-2704},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=174147},
        abstract = {Product Recommendation is a vital part of any ecommerce platform. It acts as a window for the users to surf through new products and for the sellers to market these products. It plays a crucial role in increasing sales and engagement of the ecommerce platform. These ecommerce platforms prioritize product recommendations based on sponsored listings, popularity and sales, while they do profit the organization the new age opts for recommendation of more sustainable products. Using eco-friendly products have innumerable benefits viz environmental protection, sustainable sourcing, healthier choices. Using ecommerce platforms that are considerably always in use in day-to-day life of every individual, to promote such eco-friendly products can be profitable both for the organization and the environment. Traditionally the systems would use collaborative filtering, content-based filtering to generate product recommendations. These systems have some disadvantages like cold start problem and problem in providing complex user preferences. The Large Language Models (LLM) are known to handle massive datasets of text and learn to generate human language at high level. This makes them beneficial for activities like product recommendation as they have the potential to amplify user preferences and understand product description. In this paper we propose an eco-friendly product recommendation system that uses Llama-2 LLM. Our system works by generating personalized user embedding for each user. The previous interactions with the system are taken into accord by this embedding for user preferences. The system uses this embedding to generate eco-friendly recommendations ranked highly on the list of recommended products for the user. Our results show that our system outperforms the traditional approaches on metrics like purchase rate and click-through rate. It also happens to be beneficial for promoting environmentally responsible products and creating a demand in the market for the users, eventually increasing supply of carbon-neutral products.},
        keywords = {Generative AI, Large Language Models, Ecommerce, Eco-friendly.},
        month = {March},
        }