Sri Lankan Vegetable Price Prediction System using Machine Learning

Abstract

This research aims to determine the applicability of machine learning techniques in developing a reliable, data-driven fruit and vegetable price prediction system in Sri Lanka. Data collected over the past 10 years from the Hector Kobbakaduwa Research Unit, the study accurately predicts prices for 20 key vegetables. Using the Prophet model, the system forecasts prices for today, tomorrow, and the next week. The implementation integrates a React-based front end and a NodeJS-powered backend, delivering real-time predictions through a web application. The system uses PostgreSQL as the database for storing and managing data. The Prophet model was used for machine learning purposes. The author plans to expand the system by adding more vegetables and features like comparing prices with last week's data for enhanced analysis and insights. This research makes a significant contribution to the agricultural sector by providing trustworthy price prediction model that benefits various stakeholders and aims to stabilize the agricultural market in Sri Lanka.