Winning the Misinformation Wars
Marvelous.ai is developing natural language tools to analyze political discourse in news and social media . We are particularly focused on how messaging spreads, both negatively (propaganda) and positively (pro-democracy campaigns). In this talk, we will describe our approach to detecting political narratives utilizing human-in-the-loop alongside other natural language processing techniques, with examples focused on the 2020 US presidential election. Our demo will focus on the political bias of the tweeters and asymmetries in coverage of male vs. female candidates.
- Ideas and techniques for assessing sentiment in small datasets
- Insights from a practical use case: Gender bias in politics
- Insights from a practical use case: Misinformation
Danielle Deibler is the Co-Founder and CEO of Marvelous.ai, a new startup focused on building natural language technology to discover and expose propaganda, disinformation and bias and enable advocates and policymakers to devise counter-measures and immunities. She has over 25 years in the Internet infrastructure, security, networking, interactive technology, machine learning and AI technologies. Her primary area of focus in the last 20 years has been building scalable real time interactive platforms. Previously she was CEO and Co-Founder of leading edge Reg-Tech startup Compliance.ai. She founded Apps54 and Ignited Artists. Prior to that she was an Entrepreneur in Residence at Trinity Ventures. Deibler has held senior leadership positions in software development, engineering, business development and product management for KIXEYE, Adobe, DIGEX, and UltraDNS