Skip to content

Jxk0be/YouTube-Metadata-ML-Research-Project

Repository files navigation

YouTube Metadata Machine learning Research Project | COSC526 @ UTK

Motivation

  • Assist individuals seeking success on the YouTube platform by identifying the key variables that determine a video and channel's success.
  • Utilize predictive modeling techniques, such as linear regression and decision trees to identify the critical variables that play the most significant roles in determining success in channel and vidoes.

Dataset

YouTube Dataset

Methodology

  • Data Collection - Dataset from Kaggle
  • Preprocessing - Pandas python library
  • Feature Selection - Visualizations and identifying correlations
    • Regression Techniques - Linear and Multiple Regression Tecnniques
    • Classification Techniques - Decision Trees/Random Forest from scikit-learn library
  • Evaluation - Model accuracies

Repository Structure

About

COSC 526 final machine learning project.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published