Benjamin Van Durme
- Natural Language Processing specifically computational semantics
- Streaming/Randomized Algorithms
Benjamin Van Durme is an associate professor in computer science with a secondary appointment in cognitive science. He is a member of the Center for Language and Speech Processing (CLSP) and leads natural language understanding research at Johns Hopkins University’s Human Language Technology Center of Excellence.
Van Durme’s work focuses on how computers acquire knowledge about the world, whether it is specific facts (“Johns Hopkins University is located in the city of Baltimore”), or general knowledge (“Universities are attended by students, and are sometimes located in cities”).
His research involves topics in natural language processing, data mining, social media analysis, machine learning, linguistic semantics, and other topics related to artificial intelligence and cognitive science. With his students, and with his colleagues at Johns Hopkins and other universities, he strives to teach computers how to extract structure from text, how to design algorithms and data structures for working with large collections of data, and to find ways in which machines can learn to comprehend how we convey knowledge through language – the meaning behind the words.
Van Durme received his PhD in computer science and linguistics in 2010 from the University of Rochester, where he earlier earned an MS in computer science and a BS/BA in computer science/cognitive science. He earned an MS in language technologies in 2004 from Carnegie Mellon University. At Johns Hopkins, he teaches and has taught courses in artificial intelligence, knowledge discovery in text, and event semantics in theory and practice.
Most of Van Durme’s team’s publicly-released software can be found at the JHU HLTCOE GitHub and Docker sites. His work in decompositional semantics is starting to be organized at Decomp.net, with software available.