https://github.com/feder-cr/datascienceproject

https://github.com/feder-cr/datascienceproject

Last synced: 6 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/feder-cr/datascienceproject
• Owner: feder-cr
• Created: 2024-02-11T18:13:01.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-02-15T20:58:19.000Z (almost 2 years ago)
• Last Synced: 2025-06-23T04:35:31.344Z (7 months ago)
• Language: Jupyter Notebook
• Size: 348 KB
• Stars: 6
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
University project of the course "INTRODUZIONE ALLA DATA SCIENCE" of the computer science university of Genoa

# Streaming Platform Data Analysis

## Key Features

### Objectives

- Perform comparative analysis between Netflix and Disney+ platforms 

- Apply full data science pipeline from data collection to insights

- Hands-on practice of data science concepts and techniques

### Datasets

- Disney and Netflix title datasets 

  - Metadata like title, type, date added, genre etc. 

  - Enrichment data like IMDB, TMDB scores

- 4 datasets integrated provide multifaceted view 

### Data Loading and Preparation

- Importing CSV datasets into dataframes

- Combining titles, enrichment and country datasets  

- Handling missing values and duplicate rows 

- Data transformations for analysis suitability

  - Extracting year/month from dates

  - Determining number of genres

### Exploratory Data Analysis

- Statistical summaries of key variables  

- Analysis across dimensions like certification, type etc.

- Comparative analysis across the platforms

- Hypothesis testing for distribution differences 

### Multidimensional Analysis

- Constructed OLAP cube with dimensions:

  - Month

  - Content type 

  - Production country

- Slicing, filtering and aggregation capabilities   

### Predictive Modeling

- Developed classification model using Logistic Regression

- Predict content type - movie or TV show

- Features: popularity, ratings, metadata  

- 85% accuracy on test data

### Interactive Visualizations

- Graphics for distributions, comparisons and trends

- Dashboards for slicing OLAP cube on multiple axes 

## Key Technologies

- Python

- Jupyter Notebook

- Pandas

- Numpy

- Scikit-Learn

- Matplotlib

- Seaborn

## What I Learned

Importing, cleaning and transforming medium-size datasets

Identifying optimal data formats and structures

Applying multivariate analysis techniques

Training and evaluating classification models

Using visualizations to extract insights