Hip Hop Analytica: Unveiling Social Media Dynamics in Hip Hop Discussions

Project Description:

Hip Hop Analytica is an analytical project designed to uncover and analyze the underlying structures of online interactions and discussions surrounding hip-hop topics on social media networks. This project uniquely combines community detection and sentiment analysis (specifically polarity detection) to provide comprehensive insights into the dynamics of hip-hop communities online.

I created this project in collaboration with my colleagues Kristian Perriù, Chibuzor J. Amadi, as part of the final Web and Social Networks Search and Analysis exam assignment.

Features

• Data Extraction and Processing:

  • Utilizes web scraping to collect data from Reddit, focusing on comments and interactions related to Drake's feud with Kendrick Lamar.
  • Implements data pre-processing steps, including lowercasing, stopword removal, and tokenization, to prepare the raw text for analysis.

• Sentiment Analysis:

  • Employs the advanced language model Llama 3 to conduct sentiment analysis on user comments, generating nuanced opinions on the ongoing feud.
  • Focuses on polarity detection to categorize sentiments as positive, negative, or neutral.

• Community Detection:

  • Utilizes the Louvain method and label propagation algorithms to identify distinct user communities based on their interactions and sentiments.
  • Visualizes the community structures through graphs and network diagrams, highlighting key influencers and central discussion themes.

Combining Community Detection and Sentiment Analysis

• Merges sentiment analysis results with community detection to provide a richer understanding of each community's collective stance and interactions.
• Colors nodes in visualizations based on sentiment scores to illustrate the distribution of opinions within and across communities.

Applications

• Offers insights into the social dynamics of hip-hop fandoms and the impact of social media on the evolution of hip-hop.
• Demonstrates the effectiveness of integrating community detection and polarity detection in social media analysis, with potential applications in marketing, cultural studies, and digital sociology.