Multilingual Web-based Agricultural Decision Support System: Integrating Machine Learning for Accurate Vegetable Price Prediction

This research study introduces a comprehensive approach to vegetable price prediction through a web-based application, combining Flask for the frontend and a Random Forest Regressor for robust forecasting. The dataset, encompassing vegetable attributes, seasonal factors, temperature, disaster history, and conditions, undergoes meticulous preprocessing to optimize model training. The multilingual web interface, supporting English, Telugu, and Hindi, facilitates broader accessibility via language translation using the deep_translator library. The machine learning model, trained on the refined dataset, seamlessly integrates into the Flask app for real-time predictions. The user-friendly interface enables users to input variables such as vegetable type, providing immediate predicted prices and catering to diverse users for informed decision-making. This project harmoniously integrates web development, machine learning, and language translation, offering a comprehensive solution for accurate vegetable price forecasting in the agricultural market.