A Cosine Similarity–Based Framework for Plant Recommendation Using Indian Agro- Climatic and Soil Data
DOI:
https://doi.org/10.65525/prabandhan.v1i2.8Keywords:
Recommendation System, Cosine Similarity, Smart Gardening, E-commerce, Machine LearningAbstract
A recommendation system is a filtering tool designed to predict the compatibility of items for users. In this study, the domain-specific items were plants suited to Indian regions. The Plant Recommendation System provides a list of recommended plants based on user-provided zip codes and environmental conditions such as temperature, humidity, sunlight, and rainfall. The system classifies India into six soil types and eleven zones and collects data on thirty plant species, each annotated with minimum/maximum temperature ranges, average rainfall, humidity, and sunlight needs. Cosine similarity was used to match the user inputs with the plant feature vectors. In addition to recommendations, users can add items to a wish list, add to cart, and complete a checkout process through integrated e-commerce features. The system fetches real-time location and weather data using third-party APIs. Drawbacks include a limited plant database and reliance on cosine similarity; future work will expand plant data, employ hybrid recommendation methods, and integrate IoT-based sensors for automated gardening.
Downloads
Published
How to Cite
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Prabandhan Darpan - Journal of Management Studies © 2025 by Swami Vivekananda University is licensed under Attribution-NonCommercial 4.0 International