This Jupyter notebook explores the application of machine learning techniques to recognize human activities based on smartphone sensor data. The project utilizes data from accelerometers and gyroscopes to classify activities into categories such as walking, sitting, standing, and more.
The dataset consists of recordings from 30 participants performing activities of daily living (ADL) while carrying a waist-mounted smartphone that recorded 3-axial linear acceleration and 3-axial angular velocity at a constant rate of 50Hz. The dataset includes multiple variables derived from the sensor signals such as mean, standard deviation, and magnitude of these measurements.
- k-Nearest Neighbors (kNN): Utilized for its simplicity and effectiveness in classification based on the closest training examples in the feature space.
- Support Vector Machine (SVM): Chosen for its ability to function well in high-dimensional spaces and its effectiveness in classification tasks.
- Logistic Regression: Employed for its probabilistic approach and efficiency in binary and multiclass classification settings.
- Python 3.8 or above
- scikit-learn
- pandas
- numpy
- matplotlib