Volume 7, Issue 6, June – 2022 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Rainfall Prediction using Machine Learning

Algorithms: A Comparative Analysis Approach
Yeluri Kiranmai1, Thota Kedhara Varshini 2 , Tarun Kumar Kodali3, Venigalla Sai Teja4, N. Md. Jubair Basha5
1,2,3,4
B.Tech Student, 5 Associate Professor,
CSE Department, Kallam Haranadhareddy Institute of Technology, Guntur, AP

Abstract:- Rainfall prognosis is one of the most important of both life and possessions. LSTM and ConvNet
technique to anticipate the climatic conditions across Architectures are used to anticipate the monthly average
globe. Heavy Rainfall is a major issue for meteorological rainfall. There are so many models are there to predict the
department as it is closely connected to the economy and rain fall but the models will take more time and they not
life of human beings. It is a source for natural disasters provide satisfactory predictions.
like flood and drought which are encountered by people
across the globe every year. Prompt and explicit
predictions can assist proactively reduce human and
financial loss.

The paper introduces a Rainfall prediction model

using Multi Linear Regression, Random Forest
Regression and KNN Regression. The given input data is
having numerous inputs and to anticipate the rainfall in
more specific. The Mean Squared Error (MSE), accuracy
score of training and testing, R2_Score for the
parameters used to verify the proposed model. The
comparative outcome among various algorithms is also
presented in this paper.

Keywords:- Rainfall, Prediction, Machine Learning.

I. INTODUCTION

Rainfall prediction plays vital role in Indian civilization

and human life in a great range. It is challenging duty of
meteorological department to anticipate the rainfall. It is
complicated to anticipate the rainfall accurately with
changing climatic conditions. It is challenging to predict the
Fig. 1: Existing System Architecture for Rainfall Prediction
rainfall for both summer and rainy seasons. It is important to
exactly calculate the rainfall for productive use of water III. PROPOSED SYSTEM
resources, crop productivity and pre-planning of water
structures. Knowing rainfall density plays a vital role because
heavy and irregular rainfall can have many influences like
Researchers across the globe have flourished various wrecking of crops and farms, damage of infrastructure. so a
models to anticipate the rain fall mostly using random proper prediction model is crucial for an early warning that
numbers and they are homogenous to the climate data. Even can diminished the risks to life and infrastructure and also
though many models have succeeded, but it is imperative for managing the agricultural farms in proper way. These
doing research using machine learning algorithms to get conventional methods cannot work in an efficient way so by
accurate prediction. In this project, we used Linear using machine learning techniques we can produce faultless
Regression, Random Forest and KNN to predict the annual results. We can just do it by having the previous data analysis
density of rainfall for Indian dataset. of rainfall and can predict the rainfall for the future . We can
apply many techniques according to the collected data and
II. SYSTEM ANALYSIS also, we can calculate the accuracy. Different techniques
produce different results, so it is important to choose the right
A. EXISTING SYSTEM algorithm and model it according to the requirements here we
Rainfall is one of the extreme phenomena within an are using MLR (multi linear regression), Random Forest
atmosphere system. It has a level impact on water resource
regressor and KNN regression. Feeding multiple variables
management, agriculture, and ecosystems. The heavy rainfall
and predicting the result.
mostly comes up with landslides, mudslides, floods, and
other natural disasters. The rainfall-induced disasters in each
year lead to consequential destruction and mislaying of both
life and infrastructure. Therefore, rainfall prediction is a
obstacle that needs to be solved to elude or minimize losses

IJISRT22JUN203 www.ijisrt.com 1
Volume 7, Issue 6, June – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig. 2: Proposed System Architecture for Rainfall Prediction Fig. 3: Algorithm for Rainfall Prediction

IV. METHODOLOGY The data which we have been collected will undergo for
the cleaning process. We perform splitting of the input data
In this, the study examines the use of machine learning into training and testing datasets with a desired ratio by using
in the prediction of rainfall to specifications status based on a powerful python module named Scikit-learn.
data produced. The model achieved best accuracy of using
Linear Regression. The training and testing dataset were Machine learning algorithms are evaluated using the
performed by the system from Scikit-learn module. train-test split process when they are used to generate
predictions on data that was not used to train the algorithm.
Precipitation is predicted using a predictive model. A You may differentiate the performance of different machine
good examination of the data and observation of variations in learning algorithms for the predictive modeling issue. If you
the patterns of rainfall are the initial steps in this process. have a tiny dataset or if further setup is necessary (such as if
Datasets are separated into training and testing sets before the dataset is not balanced), you should not utilize the
different machine learning and statistical techniques are technique.
applied to estimate the amount of rain. Innumerable
techniques are used to reduce the mistake. V. OUTCOMES AND DISCUSSIONS

Techniques utilized in the production of this work This paper shows the study about the results that are
include: By applying an equation to collected information, obtained by performing the above method and how they are
multiple linear regression attempts to describe the transformed by using the AI architecture.
relationship between two or more parameters and a given
answer. As far as I can see, it's nothing new than an extension
of a simple regression toward the mean.

The general formula of an multivariable linear

regression model is: y=α+β1x1+ β2x2+…+ βkxk+ε where y
is an dependent variable and x1, x2… xk are independent
variables α, β are co-efficients.

When one definite variable is not clear enough to map

the relationship between the independent and also the
variable quantity, multiple Regression can be used to
describe the relationship.

IJISRT22JUN203 www.ijisrt.com 2
Volume 7, Issue 6, June – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
After successfully uploading the dataset we can observe
the visual format of the Dataset. The Data set consist of
States, Year, Months, Annual Rainfall.

Fig. 4: Main page

The main consist Title of the project Rainfall Prediction

using Machine Learning Algorithms:a Comparative Analysis
Approach and its an image format Fig. 7: Pie chart for Rainfall Prediction

Fig. 5: Load Data

To preprocess and predict the result, primarily it is Fig. 8: Encoding & Splitting
required to need to Upload the data set in csv format.
The above shown figure is performing encodings and
splitting of dataset using Scikit-learn. As the encoding and
splitting will be performed by using a powerful Scikit-learn
module.

Fig. 6: View Data

Fig. 9: Model Training

IJISRT22JUN203 www.ijisrt.com 3
Volume 7, Issue 6, June – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[4.] Aswin S, Geetha P and Vinayakumar R, “Deep
Learning Models for the Prediction of Rainfall”,
International Conference on Communication and Signal
Processing, April 3-5, 2018, India, pp. 0657-0661.
[5.] Xianggen Gan, Lihong Chen, Dongbao Yang, Guang
Liu, “The Research of Rainfall Prediction Models
Based on Matlab Neural Network”, Proceedings of
IEEE CCIS2011, pp. 45- 48.
[6.] Sam Cramer, Michael Kampouridis, Alex A. Freitas and
Antonis Alexandridis, “Predicting Rainfall in the
Context of Rainfall Derivatives Using Genetic
Programming”, 2015 IEEE Symposium Series on
Computational Intelligence, pp. 711 – 718.
[7.] Mohini P. Darji, Vipul K. Dabhi, Harshadkumar B.
Prajapati, “Rainfall Forecasting Using Neural Network:
Fig. 10: Model Predictions
A Survey”, 2015 International Conference on Advances
The above shown figure is for performing in Computer Engineering and Applications (ICACEA)
predictions to get the required output from the given inputs. IMS Engineering College, Ghaziabad, India, pp.706 –
713
VI. CONCLUSION [8.] Sandeep Kumar Mohapatra, Anamika Upadhyay,
Channabasava Gola, “Rainfall Prediction based on 100
The estimation of rainfall is foremost for water resource years of Meteorological Data”,2017 International
management, human life and environment. It also plays vital Conference on Computing and Communication
role in agriculture production. The development of the Technologies for smart Nation, pp.162 – 166.
agriculture products is based on the amount of the rainfall So, [9.] Sankhadeep Chatterjee, Bimal Datta, Soumya Sen,
it is required to anticipate the rainfall to assist farmers in Nilanjan Dey, “Rainfall Prediction using Hybrid Neural
agriculture field. Network Approach”, 2018 2nd International Conference
on Recent Advances in Signal Processing,
The proposed methodology presented in this paper Telecommunications & Computing (SigTelCom), pp.
predicts the annual density of rainfall for the Indian dataset 67 – 72.
with regarding to multiple linear regression, random forest [10.] KangWoon Hong, Taegyu Kang “A study on Rainfall
regression, KNN regression and provides upgraded results in Prediction Based on meterological TimeSeries”, 2021
terms of accuracy score in training and testing wise, MSE twelfth International Conference on ubiquitous and
and correlation. The R2 Score for Linear Regression, Future Networks (IUFN)
Random Forest and KNN Algorithms are 0.98, 0.96 and 0.97
respectively. The Mean Squared Error (MSE) for Linear
Regression, Random Forest and KNN Algorithms are
8000.45, 39383.4 and 33456.5 respectively.

The Linear Regression algorithm showing the best

result in this model. The predicted values for the linear
regression are R2 : 0.98, MSE : 12083.66.

REFERENCES

[1.] Manojit Chattopadhyay, Surajit Chattopadhyay,

“Elucidating the role of topological pattern discovery
and support vector machine in generating predictive
models for Indian summer monsoon rainfall”,
Theoretical and Applied Climatology, pp. 1-12, July
2015, DOI: 10.1007/s00704-015-1544-5
[2.] Kumar Abhishek, Abhay Kumar, Rajeev Ranjan,
Sarthak Kumar,” A Rainfall Prediction Model using
Artificial Neural Network”, 2012 IEEE Control and
System Graduate Research Colloquium (ICSGRC
2012), pp. 82-87, 2012.
[3.] Minghui Qiu, Peilin Zhao, Ke Zhang, Jun Huang, Xing
Shi, Xiaoguang Wang, Wei Chu, “A Short-Term
Rainfall Prediction Model using Multi-Task
Convolutional Neural Networks”, IEEE International
Conference on Data Mining, pp. 395-400, 2017, DOI
10.1109/ICDM.2017.49.

IJISRT22JUN203 www.ijisrt.com 4