This repository contains the Jupyter Notebooks for my final project, which focuses on comparing the performance of various machine learning algorithms on multiple datasets. Each notebook corresponds to a specific dataset and contains both the code and in-depth analyses pertaining to the challenges posed by each dataset. This project was done in groups of two. Thank you Ananya Shekhawat for working with me on this!

The aim of this project is to leverage the machine learning skills I've developed over the semester to tackle and analyze four unique datasets. Each analysis is designed to provide insights into selecting and tuning machine learning algorithms effectively for different types of data challenges. This hands-on experience helps in understanding the practical aspects of machine learning applications, such as algorithm selection and hyper-parameter optimization.

• Anaconda environment with Python 3
• Jupyter Notebook or Visual Studio Code with the Jupyter extension installed

1. Clone or download the repository:

   git clone https://github.com/cyrilbouharb/Evaluating-Machine-Learning-Algorithms-Across-Challenging-Datasets.git
   

   or directly download the zip and unzip it.

2. Navigate to the repository directory:

   cd final_project
   

1. Open Anaconda Navigator and launch Jupyter Notebook or Visual Studio Code, or run:

   jupyter notebook
   

   or

   code .
   

   if you are using VS Code.

2. Navigate to the notebook you want to view or run within the Jupyter interface or VS Code.

3. Run the notebook:

   • Execute all cells in the notebook by clicking on Kernel > Restart & Run All in Jupyter or Run All in the command bar in VS Code.

Each notebook in this repository corresponds to a different dataset as follows:

• Problem 1: final_project_handwritten_digits.ipynb - Analysis of handwritten digits dataset.
• Problem 2: final_project_titanic.ipynb - Explorations on the Titanic survivors dataset.
• Problem 3: loan_dataset.ipynb - Study on loan default predictions.
• Problem 4: parkinsons_dataset.ipynb - Investigation into detecting Parkinson’s disease from voice recordings.
• Extra Credit #2: german_credit_dataset.ipynb - Credit scoring analysis using the German credit dataset.

• Graphs and Results: All necessary graphs and model performance metrics will be generated upon running the notebooks.
• Analysis: Look for detailed answers and algorithm comparisons at the bottom of each notebook.
• Comments: Code comments and markdown notes within the notebooks will guide you through the analysis process.

Feel free to adjust the repository URL and any other specific details to better fit your actual project setup. This README structure not only provides clear instructions on how to get the project running but also neatly organizes the content, making it more accessible for other users or evaluators.