Probabilistic Modelling of Island-Driven Parsing
نویسندگان
چکیده
Two methods for stochastically modelling bidirectionality in chart parsing are presented. A probabilistic islanddriven parser which uses such models (either isolated or in combination) has been built and tested on wide-coverage corpora. The best results are accomplished by the hybrid approaches that combine both methods.
منابع مشابه
تأثیر ساختواژهها در تجزیه وابستگی زبان فارسی
Data-driven systems can be adapted to different languages and domains easily. Using this trend in dependency parsing was lead to introduce data-driven approaches. Existence of appreciate corpora that contain sentences and theirs associated dependency trees are the only pre-requirement in data-driven approaches. Despite obtaining high accurate results for dependency parsing task in English langu...
متن کاملBreaking the barrier of context-freeness
This paper presents a generative probabilistic dependency model of parallel texts that can be used for statistical machine translation and parallel parsing. Unlike syntactic models that are based on context-free dependency grammars, the dependency model proposed in this paper is based on a sophisticated notion of dependency grammar that is capable of modelling non-projective word order and isla...
متن کاملChunking + Island-Driven Parsing = Full Parsing
We present a novel method for improving parsing performance, using a stochastic islanddriven chart parser preceded by a chunking process for identifying initial islands. Two different stochastic models have been developed for the island-driven parsing. Some experiments with nominal chunking using broad-coverage grammars derived from the Penn Treebank have been performed with remarkable results.
متن کاملData-driven Parsing using PLCFRS Data-driven Parsing using Probabilistic Linear Context-Free Rewriting Systems
This paper presents the first efficient implementation of a weighted deductive CYK parser for Probabilistic Linear Context-Free Rewriting Systems (PLCFRS). LCFRS, an extension of CFG, can describe discontinuities in a straightforward way and is therefore a natural candidate to be used for data-driven parsing. To speed up parsing, we use different context-summary estimates of parse items, some o...
متن کاملTitle Consideration on Syntactic Analyses for a Speech
SUMMARY In this paper, we compared the left-to-right & top-down parsing strategy with the island-driven & botom-up strategy by using a simulated phoneme recognizer. The both strategies adopted the beam search. The syntactic constraints were represented by a context-free grammar. The word lattice for an utterrance was generated by a word spotting algorithm from an ambiguous phoneme sequence. The...
متن کامل