EXPLANATION-BASED NATURAL-LANGUAGE ACQUISITION USING UNIVERSAL LINGUISTIC PRINCIPLES AS INNATE DOMAIN THEORY

In this paper, we propose a new natural language acquisition model (ca lled EBNLA) based on explanation-based language (EBL). To apply EBL to the natural language acquisition domain, suitable universal linguisti c principles are incorporated as domain theory. The domain theory cons ists of two parts, static and dynamic. The static part, which is assum ed to be invariant and innate to the model, includes theta theory in g overnment-binding theory and universal feature instantiation principle s in generalized phrase structure grammar. The dynamic part contains c ontext-free grammar rules as well as syntactic and thematic features o f lexicons. In parsing (problem solving), both parts work together to parse input sentences. As parsing fails, learning is triggered to enri ch and generalize the dynamic part by obeying the principles in the st atic part. By introducing EBL and the universal linguistic principles, portability of the model and learnability of knowledge in the real-wo rld natural language acquisition domain can be improved.