Adaptive Natural Language Processing

In the past decades of NLP, there has been a steady shift away from rule-based, linguistically motivated modeling towards statistical learning and the induction of unsupervised feature representations. However, natural language components used in today’s NLP pipelines are still static in the sense that their statistical model or rule-base is created once, then subsequently applied without further change. In this talk, I will motivate an adaptive approach to natural language processing, where NLP components get smarter through usage over time, following a ‘cognitive computing’ approach to natural language processing. With the help of recent research prototypes, three stages of data-driven adaptation will be illustrated: feature/resource induction, induction of processing components and continuous data-driven learning. Finally, I will discuss challenges in the evaluation of adaptive NLP components. Bio After obtaining his diploma and PhD in computer science from the University of Leipzig, Chris Biemann spent three years in the search industry at Powerset and Bing in California. Since 2011, Chris is assistant professor for language technology at TU Darmstadt, Germany. Chris and his group regularly organize shared tasks and release NLP software and data under permissive licenses. Chris’ research interests include unsupervised lexical acquisition, statistical semantics, cognitive computing and web-scale natural language processing.