Analyzing User Engagement And Predicting Popularity on Online Forums 

Project Summary

In our presentation, my team and I explored topic modeling on the UW subreddit using natural language processing (NLP). The motivation was to understand user engagement and monitor problematic content on Reddit. This approach is key in addressing the ethical challenges posed by social media, offering insights into the spread of misinformation and harmful content, and contributing to the responsible use of these platforms.

Our study focused on the UW subreddit, aiming to unravel the factors behind user engagement. We developed an NLP model to predict post popularity based on titles, using various models and sentiment analysis. This endeavor also included addressing ethical concerns related to the use of such models in online communities, with the ultimate goal of fostering a safer online environment.

We collected data from the UW subreddit using Reddit's API and analyzed it using different models and word embedding techniques. Our findings from these experiments provided valuable insights into user engagement dynamics on social media. Although our study had limitations, like focusing mainly on post titles, future research aims to broaden the dataset and incorporate more comprehensive post features to better understand engagement factors on platforms like Reddit.