https://github.com/helenaden/data-science-fundamentals

This project delves into fundamental data science concepts using Python libraries like NumPy and Pandas
https://github.com/helenaden/data-science-fundamentals

data-analysis datascience datasets datavisualization datawrangling heatmap numpy pandas patterns python

Last synced: 3 months ago
JSON representation

This project delves into fundamental data science concepts using Python libraries like NumPy and Pandas

• Host: GitHub
• URL: https://github.com/helenaden/data-science-fundamentals
• Owner: Helenaden
• Created: 2025-09-16T11:35:09.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2025-09-16T12:27:44.000Z (10 months ago)
• Last Synced: 2025-09-16T14:41:32.234Z (10 months ago)
• Topics: data-analysis, datascience, datasets, datavisualization, datawrangling, heatmap, numpy, pandas, patterns, python
• Language: Jupyter Notebook
• Homepage:
• Size: 530 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Data Science Fundamentals: NumPy & Pandas with MovieLens Case Study  

## Project Overview  

This project is all about building a solid foundation in **data science with Python**.  

Using the **MovieLens dataset**, I explored how to work with **NumPy** and **Pandas** to analyze data, uncover patterns, and draw meaningful insights.  

The dataset provides a great real-world example, combining user demographics, movie information, and ratings, the perfect playground for practicing data wrangling, analysis, and visualization.  

## Objective  

The main goal was to analyze the MovieLens datasets (movies, users, and ratings) to:  

- Understand how movies are rated and identify rating trends.  

- Explore genre preferences and user behavior.  

- Investigate the connection between demographics (age, gender, occupation) and ratings.  

## Dataset Breakdown  

### Users  

- **943 users**, each with details like age, gender, occupation, and zip code.  

- **Key findings:**  

  - The average user age is **34** (range: 7–73).  

  - Zip code values stood out as an area worth deeper investigation.  

### Movies  

- **1,680 movies** with titles, release dates, and up to 18 genre tags.  

- **Key findings:**  

  - Movies often belong to multiple genres.  

  - **Drama** and **Comedy** were the most common.  

### Ratings  

- **100,000 ratings** linked to users and movies, each with a timestamp.  

- **Key findings:**  

  - The average movie rating is **3.53** out of 5.  

## Insights & Discoveries  

- **Genre Trends:**  

  - Movies are spread across 18 genres.  

  - About half belong to more than one genre.  

  - Drama and Comedy dominate in volume.  

- **Genre Preferences:**  

  - **Film-Noir** had the highest average rating (3.92).  

  - **Fantasy** scored the lowest (3.21).  

  - Overall, 72% of genres received ratings above the global average of 3.5.  

- **Movie Favorites:**  

  - By average rating: *Great Day in Harlem, A* and *Prefontaine*.  

  - By popularity: *Star Wars* had the highest number of ratings.  

- **Demographics & Ratings:**  

  - The dataset is **71% male**.  

  - Men and women rated movies almost the same (~3.53).  

  - Non-working users gave the highest ratings.  

  - Healthcare workers gave the lowest, especially female healthcare workers.  

## Skills Applied  

- Data cleaning and preprocessing with **NumPy** and **Pandas**.  

- Exploring datasets with descriptive statistics and summaries.  

- Deriving insights from real-world data.  

- Understanding relationships between **demographics, genres, and ratings**.  

## Why This Project Matters  

This case study shows how raw data can be transformed into meaningful insights.  

It highlights:  

- How to clean and structure real-world datasets.  

- Ways to uncover hidden patterns in data.  

- The importance of combining technical skills with curiosity-driven exploration.  

Most importantly, it lays the groundwork for **more advanced machine learning and AI applications**, where understanding the data is always the first step.