Daily Weather Data Analysis Using Decision Tree Classification

This project analyzes daily weather data to classify high-humidity days using a decision tree classifier implemented with scikit-learn.

Project Description

This project utilizes a decision tree algorithm to predict whether a day will have high humidity based on morning weather sensor readings. The data is from a weather station and includes measurements like air pressure, temperature, wind speed, and rain accumulation.

Data

The data for this project is stored in a CSV file named data/data_weather.csv. This file contains weather data collected over a three-year period.

Steps

1. Import Libraries: Necessary libraries like pandas and scikit-learn are imported for data manipulation and machine learning tasks.
2. Load Data: The data_weather.csv file is loaded into a pandas DataFrame.
3. Data Exploration: The DataFrame is analyzed to understand its structure, columns, and potential missing values.
4. Data Cleaning:
   • The 'number' column, which is not required for analysis, is removed.
   • Missing values are handled using the pandas dropna function.
5. Classification Task:
   • The relative humidity at 3 pm is converted into a binary classification label (high humidity or not) based on a threshold.
6. Feature Selection: Morning sensor readings (9 am) are chosen as features to predict afternoon (3 pm) humidity.
7. Train-Test Split: The data is split into training and testing sets using train_test_split from scikit-learn.
8. Model Training: A decision tree classifier is trained on the training data using DecisionTreeClassifier from scikit-learn.
9. Prediction: The trained model is used to predict humidity labels for the testing data.
10. Evaluation: The accuracy of the model is evaluated using accuracy_score from scikit-learn.

Running the Project

1. Ensure you have Python and the required libraries (pandas, scikit-learn) installed.
2. Clone or download this repository.
3. Navigate to the project directory in your terminal.
4. Run the script using python DAILY WEATHER DATA ANALYSIS USING DECISION TREE CLASSIFICATION.ipynb.

Additional Notes

• This is a basic example of using a decision tree for weather classification.
• More advanced techniques and feature engineering could be explored for better performance.

Future Improvements

• Implement hyperparameter tuning to optimize the decision tree model.
• Explore other classification algorithms for comparison.
• Visualize the decision tree to understand its decision-making process.

I hope this README provides a clear and informative overview of the project!