Case Study: Trust Lab achieves safer social media sentiment insights with Appen

Trust Lab Making Social Media Safer Through Sentiment Analysis

Trust Lab, founded by senior Trust & Safety leaders from Google, YouTube, Reddit, and TikTok, needed a better way to understand how people around the world perceived user-generated content. Because views on offensive or acceptable content vary widely by region, platform, and audience, they turned to Appen and its Data Annotation Platform (ADAP) to gather global sentiment data on UGC.

Using Appen’s labeling platform, Trust Lab reviewed datasets covering many content types and collected millions of judgments from worldwide contributors. The resulting sentiment analysis helped Trust Lab identify high-risk and unsafe content, accounts, and transactions at scale, giving clients clearer trust and safety metrics and recommendations to improve policies and create safer online experiences.

Trust Lab

Nick Miller

Senior Data Scientist