Sociolinguistic Information Filtering Tool

Sponsoring Agency
Quantitative Scientific Solutions LLC


The proposed work will develop tools to provide users with the information and the backend support they need to reclaim control of their experience on social media. The Sociolinguistic Information Filtering Tool (SIFT) will consist of a desktop extension and associated mobile app initially based on machine learning, natural language processing and network theory techniques. Motivated by recent work understanding human information processing in the context of warning mechanisms, the project will build these tools iteratively, with humans-in-the-loop via in-depth user studies. Proposed work will evaluate and refine algorithmic approaches for account filtering and user-initiated bulk action toward automated, malicious, or otherwise unwanted accounts.