https://github.com/youssef-saaed/activity-recognition-using-various-ml-algorithms

This project involves a comprehensive comparative analysis of various machine learning models to classify activities based on a given dataset. The analysis follows a structured approach, including data exploration, model training, model evaluation, and results interpretation to identify the best performing model.
https://github.com/youssef-saaed/activity-recognition-using-various-ml-algorithms

activity-recognition comparative-analysis cross-validation data-exploration data-visualization machine-learning model-evaluation model-training neural-networks

Last synced: 4 days ago
JSON representation

This project involves a comprehensive comparative analysis of various machine learning models to classify activities based on a given dataset. The analysis follows a structured approach, including data exploration, model training, model evaluation, and results interpretation to identify the best performing model.

• Host: GitHub
• URL: https://github.com/youssef-saaed/activity-recognition-using-various-ml-algorithms
• Owner: youssef-saaed
• License: other
• Created: 2024-06-08T08:57:52.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-06-09T11:39:59.000Z (8 months ago)
• Last Synced: 2024-06-09T12:47:15.050Z (8 months ago)
• Topics: activity-recognition, comparative-analysis, cross-validation, data-exploration, data-visualization, machine-learning, model-evaluation, model-training, neural-networks
• Language: Jupyter Notebook
• Homepage:
• Size: 1.77 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Comparative Analysis of Machine Learning Models for Activity

## Overview

This project involves a comprehensive comparative analysis of various machine learning models to classify activities based on a given dataset. The analysis follows a structured approach, including data exploration, model training, model evaluation, and results interpretation to identify the best performing model.

## Project Stages

### 1. Data Exploration

- **Loading the Dataset**: Imported the dataset and conducted an initial exploration to understand its structure.

- **Exploration**: Analyzed the distribution of data and visualized various columns.

- **Correlation Analysis**: Investigated correlations between columns to understand relationships and data patterns.

- **Resampling**: Addressed the imbalance in the dataset by resampling techniques to ensure a balanced representation of classes.

### 2. Model Training

- **Models Trained**:

  - Neural Networks

  - Ridge Classifier

  - Logistic Regression

  - K-Nearest Neighbors (KNN)

  - Support Vector Machine (SVM)

- **Fine-tuning**: Utilized techniques such as k-fold cross-validation and grid search to optimize model parameters and improve performance.

### 3. Model Evaluation

- **Testing**: Evaluated the trained models on test samples.

- **Performance Metrics**: Generated confusion matrices and calculated accuracy scores to assess the performance of each model.

### 4. Results

- **Comparison**: Compared accuracies and confusion matrices of all models.

- **Best Model**: Identified Neural Networks as the best performing model based on the comparison of evaluation metrics.

### 5. Conclusion

- **Summary**: Summarized the findings and highlighted the superiority of the Neural Networks model.