Learning from Reading Syntactically Complex Biology Texts

This paper concerns learning information by reading natural language texts. The major aim is to develop representations which are understandable by a reasoning engine and can be used to answer questions. We use abduction for mapping natural language data into concise and speci c theories underlying the textual data. Techniques for automatically generating usable data representations are discussed. New techniques are proposed to get semantically correct and precise logical representations from natural language data, in particular in cases where the syntactic complexity results in fragmented logical forms.