Parisa Kordjamshidi

University of Illinois, Urbana-Champaign

 

Abstract
Understanding spatial language is important in many applications such as geographical information systems, human computer interaction or text-to-scene conversion. Due to the challenges of designing spatial ontologies, the extraction of the spatial information from natural language still has to be placed in a well-defined framework. In this talk I discuss our proposed ontology based on cognitive-linguistic spatial concepts in natural language and on multiple qualitative spatial representation and reasoning models. We bridge between spatial language and the formal spatial representation and reasoning models. To make a mapping between natural language and the spatial ontology, we propose a novel global machine learning framework for ontology population. In this framework we consider relational features and background knowledge which originates from both ontological relationships between the concepts and the structure of the spatial language. The advantage of the proposed global learning model is the scalability of the inference when learning a global model and the flexibility for automatically describing text with arbitrary semantic labels that form a structured ontological representation of its content. The learning model is evaluated on SemEval-2012 and SemEval-2013 data from the spatial role labeling task.

Bio:
Parisa Kordjamshidi is a postdoctoral researcher in KU Leuven, Belgium. She finished her PhD in July 2013 in the same university on spatial language understanding from an artificial intelligence and machine learning point of view. She has formulated the first semantic evaluation task and benchmark of “Spatial role labeling” to consider spatial semantics in SemEval2012. Her ongoing research is focused on structured machine learning and inference for natural language understanding also extended to the biomedical domain. Currently, she is a visiting scholar in UIUC, cognitive computation group (CCG).