You are on page 1of 12

WEATHER FORECASTING USING

MACHINE LEARNING

Dhrumil Darji (8744)


Maxson Koli (8748)
Tejas Vagade (8761)

Prof. Archana Lopes

Fr. Conceicao Rodrigues College of Engineering


INTRODUCTION
Weather prediction is the task of prediction of the
atmosphere at a future time and a given area. In early days,
this has been done through physical equations in which the
atmosphere is consider as fluid. Machine learning is
relatively robust and doesn't need any other physical
variables for prediction. Therefore machine learning is much
better opportunity in evolution of weather forecasting.
OBJECTIVE
To predict the Weather and Atmosphere conditions using the
previous dataset of the weather forecasting with a focus on
improving the accuracy of prediction. This will increase the
accuracy of the weather prediction and we will get accurate
results than the traditional methods. The outcome of the
predicted system gives result such as snow, rainstorm, mist,
cloudy, overcast, shady and clear.
DESIGN AND IMPLEMENTATION
The prediction system uses dataset which includes
parameters such as temperature, atmospheric pressure,
humidity, fog, dew point. This data is then separated into
train set and test set. The model is train with the help of
histogram. The classifiers used in building the model are
Logistic Regression and Random Forest.
LOGISTIC REGRESSION

Logistic regression is one of


the most popular Machine
Learning algorithms, which
comes under the Supervised
Learning technique. It is
used for predicting the
categorical dependent
variable using a given set of
independent variables.
 
RANDOM FOREST

Random Forest is a
classifier that contains a
number of decision trees
on various subsets of the
given dataset and takes the
average to improve the
predictive accuracy of that
dataset.
CODE
Logistic Regression:

Random Forest:
RESULT

1. Logistic Regression:
RESULT
2. Random Forest:
APPLICATIONS
 Air Traffic
 Marine
 Agriculture
 Private Sector
 Military Applications
CONCLUSION

Machine learning technology can provide


intelligent models, which are much simpler than
traditional physical models. They are less resource-
hungry and can easily be run on almost any
computer. Our evaluation results show that these
machine learning models can predict weather
features accurately enough to compete with
traditional models.
THANK YOU

You might also like