tailieunhanh - Báo cáo khoa học: "An Unsupervised Model for Statistically Determining Coordinate Phrase Attachment"

This paper examines the use of an unsupervised statistical model for determining the attachment of ambiguous coordinate phrases (CP) of the form n l p n2 cc n3. The model presented here is based on JAR98], an unsupervised model for determining prepositional phrase attachment. After training on unannotated 1988 Wall Street Journal text, the model performs at 72% accuracy on a development set from sections 14 through 19 of the WSJ TreeBank [MSM93]. 1 Introduction [AR98] models. | An Unsupervised Model for Statistically Determining Coordinate Phrase Attachment Miriam Goldberg Central High School Dept of Computer and Information Science 200 South 33rd Street Philadelphia PA 19104-6389 University of Pennsylvania Abstract This paper examines the use of an unsupervised statistical model for determining the attachment of ambiguous coordinate phrases CP of the form nl p n2 cc n3. The model presented here is based on AR98 an unsupervised model for determining prepositional phrase attachment. After training on unannotated 1988 Wall Street Journal text the model performs at 72 accuracy on a development set from sections 14 through 19 of the WSJ TreeBank MSM93 1 Introduction The coordinate phrase CP is a source of structural ambiguity in natural language. For example take the phrase box of chocolates and roses Roses attaches either high to box or low to chocolates . In this case attachment is high yielding H-attach box of chocolates and roses Consider then the phrase salad of lettuce and tomatoes Lettuce attaches low to tomatoes giving L-attach salad of lettuce and tomatoes Previous work has used corpus-based approaches to solve the similar problem of prepositional phrase attachment. These have included backed-off CB 95 maximum entropy RRR94 rule-based HR94 and unsupervised AR98 models. In addition to these a corpusbased model for PP-attachment SN97 has been reported that uses information from a semantic dictionary. Sparse data can be a major concern in corpusbased disambiguation. Supervised models are limited by the amount of annotated data available for training. Such a model is useful only for languages in which annotated corpora are available. Because an unsupervised model does not rely on such corpora it may be modified for use in multiple languages as in AR98 . The unsupervised model presented here trains from an unannotated version of the 1988 Wall Street Journal. After tagging and chunking the text a rough heuristic

• Báo cáo khoa học
• An Unsupervised Model
• Statistically Determining
• Coordinate Phrase Attachment
• báo cáo khoa học
• mô hình ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• A Statistical Model for Unsupervised
• Semi supervised Transliteration Mining
• Hassan Sajjad Alexander
• báo cáo ngôn ngữ
• ngôn ngữ tự nhiên
• A Bayesian Model
• Unsupervised Semantic Parsing
• Ivan Titov Saarland
• Joint Phrase Alignment and Extraction
• Graham Neubig
• Unsupervised Language Model Adaptation
• Incorporating Named Entity Information
• Long Papers
• Unsupervised Coreference Resolution
• Nonparametric Bayesian Model
• Aria Haghighi and Dan Klein
• A Language Independent Unsupervised Model
• Morphological Segmentation
• Vera Demberg
• Unsupervised Discovery
• Persian Morphemes
• Mohsen Arabsorkhi
• scientific reports
• model language
• process natural language
• Unsupervised Detection
• Downward Entailing Operators
• Maximizing Classification Certainty
• An Unsupervised Dynamic
• Bayesian Network
• Measuring Speech Style Accommodation
• Unsupervised Methods
• Head Assignments
• Federico Sangati
• Unsupervised Recognition of Literal
• Non Literal Use
• Idiomatic Expressions
• Combining Unsupervised
• Lexical Knowledge Methods
• Word Sense Disambiguation
• A Bayesian Approach
• Unsupervised Semantic
• Role Induction
• A Clustering Approach
• the Nearly Unsupervised Recognition
• Nonliteral Language
• BMC Bioinformatics
• Animal behavior
• Unsupervised learning
• Shape matching
• Shape context
• Active shape model
• Deep learning
• Glioblastoma multiforme
• Deep belief network
• Gene expression
• Model selection
• báo cáo hóa học
• công trình nghiên cứu về hóa học
• tài liệu về hóa học
• cách trình bày báo cáo
• Cellular heterogeneity
• Matrix factorization
• Particle swarm
• Quadratic programming
• Transcriptomic machinery
• Signal transduction
• Deep hierarchical neural network
• Genome Biology
• Mixture model
• Microbial species
• Statistical metagenomic algorithm
• Genomic fragments