Turkish Journal of Physiotherapy and Rehabilitation ; 32(3) ISSN 2651-4451 | e-ISSN 2651-446X www.turkjphysiotherrehabil.org 1874 Indian Crop Production:Prediction And Model Deployment Using Ml And Streamlit Arul Saxena 1 , Muskaan Dhadwal 2 , M. Kowsigan 3 1,2,3 Department of Computer Science and Engineering, College of Engineering and Technology, Faculty of Engineering and Technology, SRM Institute of Science and Technology, SRM Nagar, Kattankulathur 603203, Kanchipuram, Chennai, TN, India 1 av9152@srmist.edu.in, 2 md69082srmist.edu.in, 3, 4 kowsigam@srmist.edu.in ABSTRACT: India is an agriculture-based nation employing over 50% of the country’s workforce. However, despite being called as the backbone of India’s economy, the Industry has faced a lot of instability recently. Hence it is only natural that research of various magnitude is continuously directed towards this field in order to guarantee higher returns in the near future. Keeping in mind the many traditional approaches already prevalent in the agriculture sector which have faced numerous shortcomings, we would like to propose a competent analysis on the Indian agriculture scenario and how climate change is affecting it. This is aimed at providing help to the concerned authorities and any future research. Regression algorithms like Ridge Regression, Random Forest etc. have been used to predict the target variable (Production) and have shown great accuracy within the range of 65%-88 % depending on different algorithms. All this data is deployed on an interactive web application using Streamlit. The proposed system will benefit a lot of people as all information related to India Agriculture can be found out at one place and also the prediction of production amount leads to better planning of resources to be used. Keywords:Data science, Data Visualisation, Machine learning, Streamlit,Regression algorithms. I. INTRODUCTION 1.1. Domain and about the project With about 15 out of 20 known major climates worldwide along with the 10th largest arable soil available globally, India is home to a variety of seasonal and perennial crops and vegetation. Accounting for about 18% of India’s economy, agriculture is therefore the largest employment and revenue generating industry of the country as of date. However, with the global and national food demand on the rise due to the ever-increasing population, the agriculture sector is unable to meet the required productivity levels. As of the 2019-2020 crop year India has recorded 296.57 million tons in food grain production alone. But, with the unreliable weather patterns further burdened by the looming threat of climate change; the onset of food crises is a major concernfor the growing Indian population.These setbacks have majorly impacted the farmers in both rural and urban settings. Since each crop has its own weather requirement that needs to be fulfilled, the recent climate change has affected this dependency .In our view all this can be changed if a proper study and analysis of all the factors is done .This is where data mining and big data come into play. They play an important role in analyzing large amounts of historic climate and crop data and make effective predictions for future crop failures so that further losses incurred to both governments as well as the common man can be prevented (Kowsigan, M., et al, 2017). In recent years, the use of various Data mining and Big data Analytics tools has been on a rise world-wide(“ Kim Nari & Lee, Yang-Won, 2016”),vijay et al,2020. Researchers all around the globe are implementing various analysis techniques to find out effective solutions to various existing problem statements as well as to predict the