A Practical System for Human-like Parsing

This paper describes a human-like natural language parser called Plink. It works by parsing left-to-right through a sentence and keeping a complete representation of the partially read sentence. It does this by combining a sophisticated unification-based grammar and grammar rule selection heuristics. Plink also functions in real world applications. To do this, it must process texts that are not grammatical and does this by combining a general grammar and taking advantage of preference levels in the rule selection heuristics. It has been evaluated on two parsing metrics: Parseval [4] and a dependency based metric [14]. Plink performs well but below the state of the art. Like humans, Plink parses in linear time, and generates one interpretation that is both syntactic and semantic. It is psycholinguistically inspired.