Applied Machine Learning for Performance Analysis

This project applies machine learning techniques to analyze and classify human movement data. It focuses on three core tasks: Human Activity Recognition, Fall Detection, and Fitness Activity Classification.

Features

• Preprocessing of time-series sensor data (e.g., accelerometer, gyroscope)
• Model training using classical ML algorithms (e.g., SVM, Random Forest, k-NN)
• Evaluation with accuracy, precision, recall, and F1-score
• Visualization of activity predictions and performance metrics

Getting Started

1. Clone the repo:

   git clone https://github.com/yourusername/performance-ml.git
   cd performance-ml

2. Install dependencies:

   pip install -r requirements.txt

3. Run the scripts in src/ or explore the notebooks in notebooks/ to train and test models.

Datasets

Data can be imported from public sources like UCI HAR or custom wearable sensor data.

Results

Model performance and confusion matrices are saved in the results/ folder.