This project explores Netflix TV shows and movies (https://www.kaggle.com/datasets/victorsoeiro/netflix-tv-shows-and-movies/data) using R. It performs data cleaning, exploratory data analysis, statistical testing, and visualization on two datasets: credits.csv (cast and crew information) and titles.csv (Netflix titles metadata). The analysis focuses on genre-based IMDb score comparison, IMDb vs TMDB score distribution, and actor role frequency with outlier detection.
- git clone https://github.com/aashika116/data-analysis.git
- cd data-analysis
- Ensure you're in the R terminal and not bash, run: source("install_packages.R") OR
- manually install using: install.packages(c("dplyr","ggplot2","tidyr","ggpubr","outliers","e1071")).
Open the project in VSCode or RStudio, set the working directory to the project root and run the notebook/script inside src/ulcasestudy.ipynb.
Genre-based IMDb analysis includes cleaning genres, selecting Action and Drama, computing mean, median, standard deviation, skewness, performing Shapiro-Wilk normality tests, detecting outliers using IQR method and visualizing results using boxplots. IMDb vs TMDB comparison includes removing outliers, comparing distributions, performing paired t-tests, and plotting density graphs. Actor role analysis includes counting roles per actor, detecting outliers, and visualizing distributions using histograms.
IMDb scores vary significantly across genres. Some actors appear in unusually high numbers of titles, identified as statistical outliers. IMDb and TMDB score distributions show noticeable differences.
dplyr, ggplot2, tidyr, ggpubr, outliers, e1071
Run all scripts from the project root directory. Ensure the datasets folder remains unchanged. This project was originally developed in Kaggle and adapted for local execution.
Add interactive Shiny dashboard, expand genre analysis, perform predictive modeling, and convert to a Quarto report.
Aashika Suresh