Predicting rainfall for agriculture in India using regression
Authors
Dhyani, Sandeep
Issue Date
2020
Degree
MSc in Data Analytics
Publisher
Dublin Business School
Rights holder
Rights
Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
Abstract
In recent years the machine learning has been proven as a powerful tool for predicting the rainfall that could be useful in many sectors. This study will focus on predicting rainfall for the agriculture sector. Indian climatic conditions vary in terms of rainfall, which can be divided and observed on the basis of states. In addition, if rainfall is categorized state-wise; the constant & highest trend can be observed in the state called Meghalaya. While the lowest rainfall can be observed in both of the following states - Leh and Rajasthan. Agriculture is the crucial player in the econ- omy of India, and it is highly dependent on agriculture and forestry, which are a ected by rainfall [Krishna Kumar et al., 2004]. Disaster due to heavy rainfall like oods leads to the destruction of crops which a ects the farming sectors. If the prediction for rainfall is made by taking monthly and seasonal data of the crop into consideration; then it would be bene cial for the agriculture sector. This study will be applying the regression algorithms by di erent models, which can help in predicting the rainfall. To achieve such results, this study will be using ve various regression models and select the best one among - Mul- tiple linear regression, KNN regression, SVM(Support Vector Machine) regression, DTR(Decision tree regression), RFE(Random forest regression. The aim is to develop a model that can predict the rainfall that will help the agriculture sector, so that rainfall doesn't become a barrier for the agricultural production.