Social Network Analysis with Python

A Python project that analyzes social network data stored in JSON format. This project demonstrates data cleaning and recommendation systems similar to those used by social media platforms.

Features

1. Data Loading

Reads and parses JSON files into Python dictionaries for analysis.

2. Data Cleaning

• Removes users with missing names.
• Removes duplicate friend IDs.
• Removes duplicate page IDs.
• Generates cleaned JSON files.

3. People You May Know

Recommends users based on mutual friends, similar to Facebook's "People You May Know" feature.

4. Pages You Might Like

Recommends pages based on pages liked by users with similar interests.

---

Project Structure

Project_1/
│
├── Project_1.ipynb              # Initial JSON loading and analysis
├── data_cleaning.ipynb          # Data cleaning operations
├── People_You_May_Know.ipynb    # Mutual-friend recommendation system
├── Pages_You_Might_Like.ipynb   # Page recommendation system
│
├── data.json
├── data2.json
├── data3.json
├── cleaned_data2.json
│
├── README.md
└── .gitignore

---

Technologies Used

• Python
• Jupyter Notebook
• JSON
• Dictionaries
• Lists and Sets
• List Comprehensions

---

Recommendation Algorithms

People You May Know

Uses mutual friends to suggest new connections.

Pages You Might Like

Uses shared interests and page likes to recommend new pages.

---

Concepts Practiced

• File Handling
• JSON Parsing
• Dictionaries
• Sets
• List Comprehensions
• Sorting with Lambda Functions
• Recommendation Systems
• Data Cleaning
• Python Functions

---

Future Improvements

• Convert notebooks into Python modules.
• Add graph visualizations using NetworkX.
• Build a GUI using Streamlit.
• Add friend recommendation weights.
• Export recommendations to CSV.

---

Author

Syeda Mahnoor

Learning Data Science and Python through hands-on projects.