The members of the lab are especially interested in natural language processing and understanding, multimedia mining, machine learning and information retrieval. We study well-informed theoretical models as well as challenging applications, often empowered by big data sets. We investigate probabilistic graphical and deep learning models, with a special focus on learning with limited supervision. We have a special interest in statistical multimodal representation learning where we explore the complementarity of language and visual data. The developed technologies are, among others, applied in the domains of bioinformatics, business intelligence, e-commerce analytics, electronic message filtering, news search and mining, user generated content mining, and World Wide Web mining and search, and contribute to the field of data science in these areas. Through its collaborations LIIR connects to other disciplines including speech processing, computer vision, data mining, artificial intelligence, cognitive science, and human-computer interaction. LIIR is a member of the Leuven Brain Institute (LBI), the Leuven Center on Information and Communications Technology (LICT) and the Leuven Big Data Platform.
