The focus of my research is on semantic representations: devising ways of representing meaning in human language so that computers can work with it. My background is framed in Computer Science and Mathematics, so my approach to this task is based on applying corpus-based learning using statistical methods, usually those originating in the field of Machine Learning. Semantic representations are present at the core of many Natural Language Processing applications (machine translation, sentiment analysis, question answering, summarization...) and are a key piece in one of the major open AI problems: language understanding.
My research work is done within the project Corpus-driven induction of linguistic knowledge, funded by the Swedish Research Council. This project is an effort to combine and expand linguistic resources with corpus-driven methods.