Search for: All records

Feature-based methods are commonly used to explain model predictions, but these methods often implicitly assume that interpretable features are readily available. However, this is often not the case for high-dimensional data, and it can be hard even for domain experts to mathematically specify which features are important. Can we instead automatically extract collections or groups of features that are aligned with expert knowledge? To address this gap, we present FIX (Features Interpretable to eXperts), a benchmark for measuring how well a collection of features aligns with expert knowledge. In collaboration with domain experts, we propose FIXScore, a unified expert alignment measure applicable to diverse real-world settings across cosmology, psychology, and medicine domains in vision, language, and time series data modalities. With FIXScore, we find that popular feature-based explanation methods have poor alignment with expert-specified knowledge, highlighting the need for new methods that can better identify features interpretable to experts. 

Trust is predictive of civic cooperation and economic growth. Recently, the U.S. public has demonstrated increased partisan division and a surveyed decline in trust in institutions. There is a need to quantify individual and community levels of trust unobtrusively and at scale. Using observations of language across more than 16,000 Facebook users, along with their self-reported generalized trust score, we develop and evaluate a language-based assessment of generalized trust. We then apply the assessment to more than 1.6 billion geotagged tweets collected between 2009 and 2015 and derive estimates of trust across 2,041 U.S. counties. We find generalized trust was associated with more affiliative words (love, we, andfriends) and less angry words (hateandstupid) but only had a weak association with social words primarily driven by strong negative associations with general othering terms (“they” and “people”). At the county level, associations with the Centers for Disease Control and Prevention (CDC) and Gallup surveys suggest that people in high-trust counties were physically healthier and more satisfied with their community and their lives. Our study demonstrates that generalized trust levels can be estimated from language as a low-cost, unobtrusive method to monitor variations in trust in large populations. 

As of March 2021, the SARS-CoV-2 virus has been responsible for over 115 million cases of COVID-19 worldwide, resulting in over 2.5 million deaths. As the virus spread exponentially, so did its media coverage, resulting in a proliferation of conflicting information on social media platforms—a so-called “infodemic.” In this viewpoint, we survey past literature investigating the role of automated accounts, or “bots,” in spreading such misinformation, drawing connections to the COVID-19 pandemic. We also review strategies used by bots to spread (mis)information and examine the potential origins of bots. We conclude by conducting and presenting a secondary analysis of data sets of known bots in which we find that up to 66% of bots are discussing COVID-19. The proliferation of COVID-19 (mis)information by bots, coupled with human susceptibility to believing and sharing misinformation, may well impact the course of the pandemic.