Reproduce “Personality Detection and Analysis using Twitter Data” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Abhilash Datta, Souvic Chakraborty, Animesh Mukherjee
Researchers at the Indian Institute of Technology Kharagpur developed a dataset of 152 million tweets from 56,000 users to facilitate MBTI personality detection from Twitter. Their approach, combining diverse features and a Random Forest classifier, achieved state-of-the-art performance on this dataset.
Key insights
- MBTI personality types correlate with distinct linguistic patterns and topics expressed on Twitter, such as Analysts' tweets being harder to read and Explorers showing higher engagement.
- Each personality class exhibits unique empath features and professional preferences, like Diplomats often discussing positive emotions and art.
- User behavior on social media, including status update frequency, favorites, and listed counts, differs significantly across MBTI types.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.