Scribd
Upload
6 views13 pages

Nnachaal@gmail - Com WeatherAnalysis

The document analyzes weather patterns using a dataset of 730 records with key features including temperature, humidity, and pressure. It employs K-NN classification and K-means clustering to predict rain presence and identify seasonal weather trends, achieving an overall accuracy of 98.6%. The findings highlight the relationship between temperature, humidity, and rain presence, while also addressing challenges such as data inconsistencies and the need for advanced analytical tools.

Uploaded by

nachaaln
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views13 pages

Nnachaal@gmail - Com WeatherAnalysis

The document analyzes weather patterns using a dataset of 730 records with key features including temperature, humidity, and pressure. It employs K-NN classification and K-means clustering to predict rain presence and identify seasonal weather trends, achieving an overall accuracy of 98.6%. The findings highlight the relationship between temperature, humidity, and rain presence, while also addressing challenges such as data inconsistencies and the need for advanced analytical tools.

Uploaded by

nachaaln
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Weather Patterns

Analysis and Prediction


NACHAAL N
20.12.2024
TABLE OF CONTENTS

01 02 03
DATA
INTRODUCTION ANALYSIS CONCLUSION
Overview of data Key findings Linking data
PROJECT OVERVIEW
● Number of Records: 730
● Number of Features: 9
● Key Columns: Date,
Weather_Condition,
Dew_Point (°C), Humidity
(%), Pressure
(hPa),Temperature (°C),
Visibility (km)
Data Analysis - FINDINGS

Maximum
Temperature
45.0°C
Minimum
Temperature:
12.0°C
Average
Humidity:
36.34%
TRENDS
Temperature Humidity
Rising temperatures High humidity
observed over time levels are more
frequent
K-NN CLASSIFICATION

Distance calculation Predicting Rain


(Euclidean). Presence.

Selecting nearest For each test point, the


neighbors. algorithm predicts the
outcome
K-MEANS CLUSTERING

Initialize
clusters (e.g., Recompute
random points) centroids.
02 04

01 Assign records 03 Final clusters


to nearest identified and
clusters. Updated centroids
calculated
K-NN
● Precision : For class 0 (No Rain): 98.6%

01
● For class 1 (Rain): 100%
● Recall:
● For class 0: 100%
● For class 1: 50%
● F1-Score:
● For class 0: 99.3%
● For class 1: 66.7%
● Overall Accuracy: 98.6%
K- MEANS CLUSTERING
A scatterplot of Temperature vs
Humidity shows:
Data points colored by cluster.
Black crosses marking the centroids.
The centroid coordinates for each cluster are as follows:

Cluster Sizes:
Cluster 1: 356 data points.
Cluster 0: 304 data points.
Cluster 2: 70 data points.
INSIGHTS AND LEARNINGS

PATTERNS RESULTS TOOLS

Rain Presence linked Clustering revealed Machine Learning,


to temperature and seasonal weather Excel
humidity groups
Challenges and
Recommendations
Handling missing or inconsistent
data.

Long Calculations for KNN

Recommend to use more


advanced softwares for deeper
analysis
Gather additional data for
improved predictions
CONCLUSION

Demonstrated the
Analyzed weather impact of data
data for Rain science in weather
Presence prediction. analysis

Used clustering to Tools used :


identify seasonal Python , Excel,
weather patterns Google and
Weather Data
CSV

You might also like