Scribd
Upload
324 views

Data Science in Meteorology - Weather Forecast: S.Y. Mechanical Department

The document discusses how data science is used in weather forecasting. It describes how weather data is collected from satellite imagery and sensors and used with machine learning algorithms like decision trees, K-means clustering, and K-nearest neighbors for classification and regression to predict weather patterns. Accurate weather forecasting helps industries like agriculture and sports scheduling. While forecasts have improved, weather remains difficult to predict due to complex variables, and forecasts still have limitations.

Uploaded by

cap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
324 views

Data Science in Meteorology - Weather Forecast: S.Y. Mechanical Department

The document discusses how data science is used in weather forecasting. It describes how weather data is collected from satellite imagery and sensors and used with machine learning algorithms like decision trees, K-means clustering, and K-nearest neighbors for classification and regression to predict weather patterns. Accurate weather forecasting helps industries like agriculture and sports scheduling. While forecasts have improved, weather remains difficult to predict due to complex variables, and forecasts still have limitations.

Uploaded by

cap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 25

S.Y.

Mechanical Department

Data Science in
Meteorology
-Weather Forecast
GROUP MEMBERS

Tushar Patil Soham Phadtare Rafe Khan Prathmesh Rajput Prasad Ruikar
51 55 61 62 75

Group - 5
 Weather Forecasting

 Brief about Water-Cycle

 Role of Data Science

Table of Contents  Datasets , Various Steps & Algorithms

 Outputs observed

 Applications & Limitations

 Conclusion
What is Weather Forecast?

Why it is important ?

How does it helps ?


Let us Study in short about change in weather
via
Water Cycle

Water never leaves the Earth. It is constantly


being cycled through the atmosphere, ocean
and land. This process, known as the water and
land.

Water Cycle , results into


change in weather
conditions.
Data Science
Behind Weather Prediction

Predective
modelling and
Machine
learning

Satellite Right Data for


imagery and Accurate
Sensor Data Decision Making

Weather Data
acts as a
support for
many events
Dataset

Some Independent Variable’s Examples used are :


Max_Temperature , Min_Temperature
WindSpeed3pm, WindSpeed9am , WindSpeed3am
Humidity9am,Humidity3pm
Rain Today, Rain Tomorrow

7
ALGORITHMS

• A Decision Tree is an algorithm used for


supervised learning problems such
Decision tree
as classification or regression
-R part

• Computes the conditional a-posterior


probabilities of a categorical class variable
K-means given independent predictor variables using
Cluster the Bayes rule.

• A typical version of K-means algorithm works


mainly with 3 steps like initial cluster
K-nearest seeds,eucledian squared distance & centroid
neighbours computation.
CLASSIFICATION & REGRESSION
TREE
 The decision trees produced by CART are strictly binary,
containing exactly two branches for each decision node.
 Formula = Rain_Tomorrow ~ min_temp+
max_temp+windspeed9am+windspeed3pm+h
umidity3pm+pressure3pm

9
DECESION TREE
• Decision Tree algorithm belongs to the family of supervised learning
algorithms.

• Unlike other supervised learning algorithms, the decision tree


algorithm can be used for solving regression and classification
problems too.
K- means Cluster
Algorithm

-Kmeans algorithm is an iterative algorithm that tries to partition


the dataset into Kpre-defined distinct non-overlapping subgroups
(clusters) where each data point belongs to only one group.
Steps involved

Choose the
number from
cluster k

Select k
Repeat step 3 random points
& 4 if needed from data as
centroids

Recompute
the centroids
Assign all
of newely
points to
formed
closest cluster
clusters using
cell
sq.Eucledian
distance
SAMPLE OUTPUTS OF K-MEANS CLUSTER METHOD

Step - 2 Step -3
Step-1
Application of Cluster Method

Market
segmention

Computer
Agriculture
Applications Vision

of Cluster
Method

Astronomy Geostatics
K- Nearest Neighbor
Method

• K nearest neighbors (KNN) is a supervised machine learning


algorithm

Formula :
Steps Involved

Receive an
unclassified
Data

Measure the
Classify new
Euclidian
Data
distance

Chose one Obtain the


from most smaller
appearance distance

Check list
for short
distance
Sample Output of KNN method
Applications of K-nearest
neighbours

Use for
Recommendation
System

Applications
of
KNN
Can effectively
Can Search
used for
similar
detecting
documents
outliers
Comparison of Algorithms
In terms of Accuracy

Decision Tree – 84%

K-means – 80.5%

KNN – 79%
Advantages of Weather Forecast

• Prior Warnings of Weather

• To take Proper Decisions

• Better Organization

• Delivers Preffered visual forecast which proves helpful

• Most helpful in agriculture sector

• Best method for Management of Inventory

• Provides Helpful Decisions


Weather Forecast
-An aid for many events

Predictions of Floods and Natural


Disasters

Sports – Organising matches


becomes efficient

Predict Asthma Attacks- chances of


attacks by predicting where asthma
can be triggered.
Limitations
Of weather forecast

• Weather is extremely difficult to forecast


correctly .

• It is expensive to monitor-so many variables


from so many sources .

• The computers needed to perform the


millions of calculations necessary are
expensive .

• The weather forecasters get blamed


if the weather is different from the
forecast .
Conclusions

• Weather forecasting is a really difficult task , The


machine learning algorithms can help in
prediction for a short term period.

• Weather forecasts are increasingly accurate and


useful, and their benefits extend widely across
the economy. 

• The forecasting community is working closely


with multiple stakeholders to ensure that
forecasts and warnings meet their specific
needs.
References
• https://towardsdatascience.com/weather-forecasting-with-data-science-
approaches-cb8f2afd3f38

• https://www.ametsoc.org/index.cfm/ams/about-ams/ams-
statements/statements-of-the-ams-in-force/weather-analysis-and-
forecasting/

• https://data-flair.training/blogs/data-science-for-weather-
prediction/
Thank You

You might also like