Biologically plausible connectionist prediction of natural language thematic relations

In Natural Language Processing (NLP) symbolic systems, several linguistic phenomena, for instance, the thematic role relationships between sentence constituents, such as AGENT and PATIENT, can be accounted for by the employment of a rule-based grammar. Another approach to NLP concerns the use of the connectionist model, which has the benefits of learning, generalization and fault tolerance, among others. Inspired on neuroscience, it is proposed a connectionist system called BIOθPRED (BIOlogically plausible thematic (θ) PREDictor), designed to reveal the thematic grid assigned to a sentence. Its architecture comprises, as input, a featural representation of the words (based on the verb/noun WordNet classification and on the classical semantic microfeature representation), and, as output, the thematic grid assigned to the sentence. BIOθPRED is designed to “predict” thematic (semantic) roles assigned to words in a sentence context, employing biologically inspired training algorithm and architecture, and adopting a psycholinguistic view of thematic theory.