Social Media Data Analysis

Project source can be downloaded from https://github.com/Ztrimus/social-media-data-analysis.git

Author & Contributor List

Saurabh Zinjad

All other known bugs and fixes can be sent to "[email protected]" with the subject "Social Media Data Analysis". Reported bugs/fixes will be submitted to correction.

Introduction:

Through this project, you will develop skills in social media data crawling, how to operate an open-source large language model while crafting compelling prompts, and performing exploratory analysis on the gathered data. For this project, you are required to crawl data from the Mastodon social media platform. Additionally, you have the freedom to choose a topic of your choice. However, it is crucial to select a controversial topic that fosters the emergence of contrasting viewpoints, such as Climate Change (pro-climate change, neutral, anti-climate change), Threads (pro-Threads, neutral, anti-Threads), Twitter (anti-Elon, neutral, pro-Elon), AI tech industry (AI tech optimist, neutral, AI tech pessimist), or other similarly engaging issues.

Get Started

1. First, Go thorugh "Project 1 - Social Media Data Analysis" pdf.
2. Obtaining Mastodon API credentials, as per mentioned in pdf.
3. Create credentials.py file and add Client Id, Secret and access token.

   MASTODON_SOCIAL_ID = "add your MASTODON_SOCIAL_ID here"
   MASTODON_SOCIAL_SECRET = "add your MASTODON_SOCIAL_SECRET here"
   MASTODON_SOCIAL_TOKEN = "add your MASTODON_SOCIAL_TOKEN here"
   API_BASE_URL = "add your API_BASE_URL here e.g. https://mastodon.social"

4. Get access of Llama2-7B model as per mentioned in pdf

Resources

• https://mastodonpy.readthedocs.io/en/1.8.1/

Bugs :

Limitation and Further growth :