Scribd
Upload
15 views

21 - Rainfall Prediction Using Machine Learning

Uploaded by

ecehod.sdgi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
15 views

21 - Rainfall Prediction Using Machine Learning

Uploaded by

ecehod.sdgi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

International Journal of Innovative Research in Engineering & Management (IJIREM)

ISSN: 2350-0557 Volume-5, Issue-6, November 2018


www.ijirem.org

Rainfall Prediction Using Machine Learning


Abhishek Prakhar Akash Tripathi Pradhyumn Srivastava
Department of Computer Science Department of Computer Science Department of Computer Science
& Engg, G L Bajaj Institute of & Engg, G L Bajaj Institute of & Engg, G L Bajaj Institute of
Technology & Management, Technology & Management, Technology & Management,
Greater Noida, India Greater Noida, India Greater Noida, India

Ujjwal Gupta
Department of Computer Science & Engg,
G L Bajaj Institute of Technology & Management,
Greater Noida, India

ABSTRACT Due to the atmospheric process rainfall creates complications


Rainfall prediction is challenging tasks in advanced world. in understanding of model in water cycle and has variation in
Rainfall and climate is high non linearly and complex process, atmosphere and lithosphere with time. This paper contains
for their exact prediction it required modern computer prediction of rainfall in wide terminology that will last. This
simulation and modeling. It providing a research and survery of research paper will provide the information about rainfall that
some methodologies on available literature to utilize various will help in its prediction of particular area. The former data will
regression technique for prediction employed by many be helping us in this in prediction of rainfall. There is already a
researchers. This model states which ‘rainfall prediction’ technique given by Holmstrom that forecast minimum and
algorithm is suitable than outdated methods. maximum temperature of next week with the help of
information of past two days. With the help of a functional
linear regression model and a linear regression model, they
Keywords proved that both of the above models were not good as
Machine Learning, Neural Networks, Data visualization professional weather forecasting services for the prediction up to
a week. However, in later days or longer time scales their model
1. INTRODUCTION performs better in forecasting. It was studied a hybrid manner,
The weather condition around the world changes rapidly and which combined many trained prediction model with the deep
continuously. Correct forecasts are essential in today’s daily life. neural networks.
From agriculture to industry, from traveling to daily commuting,
we are dependent on weather forecasts heavily. As the entire 2. OBJECTIVE AND SCOPE
world is suffering from the continuous climate change and its Rainfall Prediction Model has a main objective in prediction of
side effects, it is very important to predict the weather without the amount of rain in a specific well or division in advance by
any error to ensure easy and seamless mobility, as well as safe using various regression technique and find out which one is
day to day operations. The project predicts the weather of future best for rainfall prediction. This model also helps the farmer for
dates based on the previous data using concepts of machine agriculture to decide the crop, helping the watershed department
learning. for water storage and also helps to analyze the ground water
We take ‘Regression model’ data of Delhi’s climate from level.
1997 to 2017 as the feed and predicts the temperature of
particular date. The linear regression model is used in many
project. We are using linear regression, Random Forest and 3. METHODOLOGY
Decision Tree to predict the output, in order to get the results For conducting the experiments in correct format data is
and compare the models for its accuracy. Physically simulating altered.
the atmosphere as a fluid, weather forecast is always perform.  Observe variation in rain pattern and have good
We are sampling current atmosphere state. For solving the analysis.
numerical equations of thermo and fluid dynamics, the future  For prediction of average rainfall separated data into
state of the air is computed. In different kinds of weather, in testing and training. We apply ML tactics in pursuit
human life rainfall play the important part. Rainfall have vital of make analyzing over various methods. So it
role in human civilization. As rainfall is nature`s call, it is very reduces the error
hard to know its situation, that is why it have much demand. It is
very important to predict rainfall because it is directly linked to
the conservation of water resources, operations of the reservoir
and in detection of flood level in nearby body. Water level due
to rainfall affects the human activities such as sewer and traffic.

Copyright © 2018. Innovative Research Publications. All Rights Reserve 260


Rainfall Prediction Using Machine Learning

4. RESULTS & DISCUSSION


Table 1: Training on complete dataset
Algorithm MAE
Linear Regression 57.08862331011229
SVR 116.60671510825178
Artificial neural nets 42.132246471373314

• Neural Networks have good outcome than SVR etc.


• SVR gives Low performance.
• It gives sufficient means and SD.

5. CONCLUSION
The ways for predicting rainfall is due to many data
visualization.
• For either types of data for its prediction.
• Dataset1-This data is having mean rainfall from
1951-2000 for each district, for every month.
• Dataset2- From 1901-2015 onwards it have average
rainfall for every state.

REFERENCES
[1] ADALINE system to weather forecasting, Technical
Report, Stanford Electron, 1964
[2] Lee S. Cho S.& Wong P.M. “Rainfall prediction
using artificial neural network. Geog. Inf. Decision
Anal. 2, 233–242 1998.
[3] Michaelides, S. C., Neocleous, C. C. & Schizas, C. N.
“Artificial neural networks and multiple linear
regression in estimating missing rainfall data.” In
Proceedings of the DSP95 International Conference
on Digital Signal Processing, Limassol, Cyprus. pp
668–6731995.

Copyright © 2018. Innovative Research Publications. All Rights Reserve 261

You might also like