Skip to content

Latest commit

 

History

History
57 lines (30 loc) · 1.97 KB

README.md

File metadata and controls

57 lines (30 loc) · 1.97 KB

Social Network Analysis on Mongolia-related Tweets

About the project

I have been collecting certain tweets related to Mongolia as a topic since 2020.

Using this data, I tried to analyze and identify the most influencial users with the learnings from the class "МТССНДАМСС".

Main results and methods used

1. Reading and analyzing initial data with Pandas and Numpy

1. Reading and analyzing initial data with Pandas and Numpy

2. Preparing and minimizing data with Pandas for limited processing time

2. Preparing and minimizing data with Pandas for limited processing time

3. Creating a graph with trios of (retweeter, retweetee, tweet_count) with Networkx

3. Creating a graph with trios of (retweeter, retweetee, tweet_count) with Networkx

4. Calculating betweenness and other centralities with Networkx

4. Calculating betweenness and other centralities with Networkx

5. Visualizing the graph with Matplotlib

5. Visualizing the graph with Matplotlib

6. Modifying the graph and adding visual aids with Networkx and Matplotlib

6. Modifying the graph and adding visual aids with Networkx and Matplotlib

To replicate the results

Requirements

pip install -r requirements.txt

After preparing the environment, please start the jupyter notebook sna-mongolia-tweets.ipynb and follow the code steps.

Data

The tweet dataset I used is part of an ongoing data collection effort by me.

All data is included in this repo to be openly used.

Code

Initial data analysis and data preparation code was written from scratch by me.

Pandas official documentation and Stackoverflow were used, of course.

Core of the Social network analysis code was taken from this project by Julia Wu.