This project focuses on analyzing Netflix user watching behavior using data preprocessing, statistical analysis, visualization, and feature engineering techniques.
The analysis helps identify patterns related to user engagement, subscription behavior, binge watching activity, and churn trends.
Source: https://www.kaggle.com/datasets/rhythmghai/netflix-user-watching-behavior-dataset
The dataset contains user demographics, subscription details, engagement metrics, viewing activity, and interaction patterns.
- Python
- Pandas
- NumPy
- Matplotlib
- Seaborn
- SciPy
- Scikit-learn
- Jupyter Notebook
- Data Cleaning
- Missing Value Analysis
- Outlier Detection
- Scaling and Encoding
- Exploratory Data Analysis
- Statistical Testing
- Correlation Analysis
- Feature Engineering
- Visualization
- Pearson Correlation
- T-Test
- ANOVA
- Chi-Square Test
- Shapiro-Wilk Test
- QQ Plot Analysis
New features created:
- Engagement Score
- Binge Watcher Classification
- Active User Classification
- User engagement patterns vary across subscription types.
- Binge watching behavior is common among active users.
- Statistical tests helped identify relationships between subscription plans and churn behavior.
- Feature engineering improved behavioral interpretation of users.
Saras Kumar
M.Sc. Data Science & Analytics
Ramaiah University of Applied Sciences