The Sims 4 Community Analysis: Sentiment, Topics & Player Experience Insights

This project combines data science, NLP, and product thinking to analyze how The Sims 4 player community has evolved over the past decade. By processing 591K+ Reddit posts and comments, the analysis reveals long-term sentiment trends and key topics that drive player satisfaction, frustration, and engagement.

Project Overview

Player communities generate enormous volumes of unstructured feedback. This project demonstrates how scalable NLP pipelines can turn that data into actionable insights for both research and real-world product improvement.

It integrates sarcasm-aware sentiment analysis, BERTopic modeling, and SQL-based data engineering to extract interpretable patterns from noisy social media text — all while maintaining research transparency and reproducibility.

Key Highlights

• Dataset: 591,584 Reddit posts and comments (2014–2024) from Sims-related subreddits
• Tech Stack: Python, pandas, spaCy, NLTK, VADER, BERTopic, SQL
• Approach: Text cleaning, sarcasm detection, sentiment classification, and topic modeling
• Outputs:
  • 42 discussion clusters condensed into 5 high-impact opportunity areas
  • Year-over-year sentiment tracking for community health analysis
  • Reproducible notebook pipeline for future topic trend updates

Key Findings

• Gameplay depth and AI behavior consistently rank among the most discussed (and negatively scored) topics.
• Sentiment toward representation and diversity has trended positive since 2020, though engagement has plateaued.
• Incorporating sarcasm-sensitive models improved classification accuracy by 8% compared to standard sentiment tools.

You can try this project out yourself at THIS link.

Future Directions

• Expand to full Reddit comment trees (≈600K+ additional records)
• Integrate time-aware embeddings for longitudinal topic sentiment tracking
• Compare BERTopic results to transformer-based topic models (e.g., BERTopic + BERTopic-CTFIDF hybrid)

---

Author: Bader Rezek
*University of Illinois Chicago | Data Science