Modelling Relations in Texts and Knowledge Graphs for Logical Reasoning & Graph-Text Conversion.

Martin Schmitt is a senior PhD student at CIS (Center for Information and Language Processing), LMU Munich, supervised by Prof. Dr. Hinrich Schütze. In his research, he combines natural language processing with knowledge graphs. Besides extracting knowledge from texts and generating textual summaries from knowledge bases, his main interest lies in lexical semantics and its use in natural language Inference. Abstract: Knowledge graphs (KGs) store facts in the form of triples that contain two entities and the relation between them. Typically this relation is assigned a short phrase, such as “born in“, to describe its meaning intuitively. In this way, the triple resembles a natural language statement although it often lacks the fluency of a real sentence. It is common to reason about facts from KGs and texts to infer new knowledge. An important yet underexplored setting for this is relation inference in context (RIC). Here, the validity of a triple (e_1, r, e_2)