Computational Cognitive Linguistics

Computational Cognitive Linguistics Jerome A. Feldman ICSI & UC Berkeley 1947 Center Street Berkeley, CA, 94704 jfeldman@icsi.berkeley.edu Abstract 1 The talk will describe an ongoing project (modestly named the Neural Theory of Language) that is attempting to model language behavior in a way that is both neurally plausible and computationally practical. The cornerstone of the effort is a formalism called Embodied Construction Grammar (ECG). I will describe the formalism, a robust semantic parser based on it, and a variety of applications of moderate scale. These include a system for understanding the (probabilistic and metaphorical) implications of news stories, and the first cognitively plausible model of how children learn grammar. 2 Introduction Computational Linguistics is sometimes narrowly identified with statistical corpus studies, but is much broader. The learning and use of language are inherently information processing tasks and any systematic description of them must be computational. If we want to understand how human brains acquire and exploit language, we are driven to computational theories and models that link neural structure to linguistic behavior. Cognitive Linguistics provides the basic mechanisms for explaining human language processing, but has traditionally been informal and non-computational. Recent work suggests that a formally grounded computational cognitve linguistics can be quite productive. 3 The Neural Theory of Language For some sixteen years, an interdisciplinary group at ICSI and UC Berkeley has been building a theory of language that respects all experimental findings and is consistent with biological and computational constraints. The intellectual base for the project is a synthesis of cognitive linguistics and structured connectionist modelling. This involves work from neurobiology through politics and philosophy (Lakoff 1999). An accessible overview of the goals and methodology of the project can be found in (Regier 1996) and further information is available at: WWW.icsi.berkeley.edu/NTL The focus of this talk is computational; how can an embodied theory of language support scalable systems for language learning and use. The key to scalability in any paradigm is compositionality; our goal in modeling language understanding is to systematically combine the heterogeneous structures posited in cognitive linguistics to yield overall interpretations. We have identified four conceptual primitives that appear to capture the suffice for building scalable language understanding systems: SCHEMA, MAP, (MENTAL) SPACE, and CONSTRUCTION. Schemas are language-independent representation of meanings and constructions are form-meaning pairs. Constructions are form-meaning mappings as depicted in Figure 1. Maps and mental spaces are discussed in (Mok 2004). The ECG formalism has, in addition to the usual inheritance hierarchy, an EVOKES relation that makes an outside structure accessible to a schema through a local name. As shown in Figure 1, langauge understanding is implemented as having distinct analysis and simulation phases. During analysis, a Semantic Specification (SemSpec) is created from the meaning poles of the constructions, and is essentially a network of schemas with the appropriate roles filled in.. Unification of constructions requires compatability of their embodied semantic scehemas as well as form matching. Crucially, within this network of schemas are executing schemas (or x-schemas Narayanan 1999), which are models …

Content truncated to 500 words