Foreseer Group

Welcome to the Foreseer Group! We are a research team within the University of Michigan School of Information (UMSI). We work on cutting-edge research related to data mining, machine learning, information retrieval, and natural language processing. We build models, algorithms, and systems to analyze large collections of text, networks, and behaviors. These techniques are applied to solving problems in web mining, social computing, health informatics, scientific research, and other domains with the aim to generate a positive impact on everyone’s daily life. Our research is supported by various government agencies such as NSF, NIH, and DARPA. We collaborate with various industry leaders and academic partners.

Our recent interests include, but are not limited to:

• AI for Behavioral Science ([KDD Workshop on AI Behavioral Science], [Turing Test for AI Chatbots, PNAS])
• LLMs for Scientific Discovery ([MASSW], [SciLitLLM])
• Machine Learning on Graph Structures ([LLMs on Graph Structures, TMLR], [Hypergraph Expansion, WWW 2024], [A Metadata-Driven Approach for GNNs, NeurIPS 2024], [PM2.5 Forecasting, AI Open])
• AI for Social Science ([AI for Crowdfunding], [Team Resilience during Covid, AAAI 2023])

Talk to us or join our group if you are interested in working on these exciting research topics! We are always looking for talented students to join our group.