Development Of AI-MI Based Models for Predicting Prices of Agri-Horticultural Commodities
Puneeth N
Department of Computer Science and Engineering,
Presidency University, Bangalore, Karnataka, India
puneethpunee104@gmail.com
Rohan Gowda A
Department of Computer Science and Engineering,
Presidency University, Bangalore, Karnataka, India
a.rohan1998@gmail.com
Thanu Shree M
Department of Computer Science and Engineering,
Presidency University, Bangalore, Karnataka, India
thanushreemanjunath03@gmail.com
Ananya B R
Department of Computer Science and Engineering,
Presidency University, Bangalore, Karnataka, India
ananyabrbr@gmail.com
Dr. Shanthi S
Associate Professor
Department of Computer Science and Engineering,
Presidency University,
Bangalore, Karnataka, India
shanthi.s@presidencyuniversity.in
Ajin V Joseph
Department of Computer Science and Engineering,
Presidency University, Bangalore, Karnataka, India
ajinvarghese188@gmail.com
Abstract : The Vegetable Price Prediction project aims to develop a reliable predictive model to predict vegetable prices based on crucial seasonal, weather, and crop health data. As vegetable price is a top concern for farmers, suppliers, and consumers with volatile vegetable prices, accurate price predictions can facilitate informed decision-making for stakeholders. This project makes use of historical data like vegetable kind, season of growth, month, temperature, recent calamity occurrences, state of vegetable, and other variables that affect the price per kilogram to predict the price per kilogram. Utilizing advanced machine learning algorithms ARIMA, LSTM, XGBoost, Linear Regression and Hybrid Algorithms we aim to identify complex patterns and time dependencies in the data to make the prediction. Each algorithm has a strength: ARIMA for time series, LSTM for long-range dependencies, XGBoost for performance, and Linear Regression for simplicity and interpretability. The results of the model will be verified for effectiveness and accuracy, making strong predictions in different scenarios.
Keywords: Machine Learning Models, ARIMA, LSTM (Long Short-Term Memory), XGBoost, Linear Regression, Hybrid Algorithms.
