Machine-learning-Assignment01

Predicting heart disease with decision tree classification

Decision Tree Classification and Analysis

This project explores the application of decision tree classification on a given dataset. It includes data analysis, model building, evaluation, and visualization.

Project Description

This project performs the following tasks:

1. Data Analysis:
   • Exploratory data analysis (EDA) using Pandas to understand the dataset's characteristics.
   • Visualization of data distributions using Matplotlib and Seaborn (e.g., histograms, bar charts, pie charts).
   • Duplicate data handling.
2. Model Building:
   • Building a decision tree classifier using Scikit-learn's DecisionTreeClassifier.
   • Splitting the data into training and testing sets.
   • Hyperparameter tuning using GridSearchCV to optimize the model and mitigate overfitting.
3. Model Evaluation:
   • Evaluating the model's performance using accuracy scores.
   • Generating and visualizing a confusion matrix.
   • Generating a classification report with precision, recall, and F1-scores.
   • Analyzing feature importance.
4. Tree Visualization:
   • Visualizing the decision tree using sklearn.tree.plot_tree for basic visualization.
   • Visualizing the decision tree using export_graphviz and graphviz for more detailed and customizable visualization.

How to Use

1. Clone the Repository:

   git clone [https://github.com/Viraj97-SL/Machine-learning-Assignment01.git]
   cd [https://github.com/Viraj97-SL/Machine-learning-Assignment01.git]

2. Install Dependencies:
   • Create a virtual environment (recommended):

     python -m venv venv
     source venv/bin/activate  # On macOS/Linux
     venv\Scripts\activate      # On Windows

   • Install the required Python libraries:

     pip install pandas scikit-learn matplotlib seaborn graphviz

   • Install Graphviz software:
     • You must also install the Graphviz software on your system.
     • Windows: Download the installer from http://www.graphviz.org/download/ and add the bin directory to your system's PATH.
     • macOS: brew install graphviz
     • Linux (Debian/Ubuntu): sudo apt-get install graphviz
     • Linux (Fedora/CentOS): sudo yum install graphviz
3. Run the Jupyter Notebook:
   • Start Jupyter Notebook:

     jupyter notebook

   • Open the .ipynb file and execute the cells.

Dependencies

• Python Libraries:

  • pandas (for data manipulation)
  • scikit-learn (for machine learning)
  • matplotlib (for basic plotting)
  • seaborn (for advanced plotting)
  • graphviz (for decision tree visualization)

• Graphviz Software:

  • Graphviz is required for the advanced tree visualization. You need to install it separately according to your operating system.

Files

• Assignment(ML&NN).ipynb: Jupyter Notebook containing the code and analysis.
• hearts.csv : The dataset used in the analysis.
• README.md: This file.

Notes

• Ensure that the Graphviz bin directory is added to your system's PATH environment variable for the graphviz visualization to work correctly.
• The notebook assumes that the data file is in the same directory as the notebook. Update the file paths if necessary.
• Feel free to modify the code and parameters to experiment with different settings and datasets.