TAILIEUCHUNG - Báo cáo khoa học: "Models and Training for Unsupervised Preposition Sense Disambiguation"

We present a preliminary study on unsupervised preposition sense disambiguation (PSD), comparing different models and training techniques (EM, MAP-EM with L0 norm, Bayesian inference using Gibbs sampling). To our knowledge, this is the first attempt at unsupervised preposition sense disambiguation. | Models and Training for Unsupervised Preposition Sense Disambiguation Dirk Hovy and Ashish Vaswani and Stephen Tratz and David Chiang and Eduard Hovy Information Sciences Institute University of Southern California 4676 Admiralty Way Marina del Rey CA 90292 dirkh avaswani stratz chiang hovy @ Abstract We present a preliminary study on unsupervised preposition sense disambiguation PSD comparing different models and training techniques EM MAP-EM with L0 norm Bayesian inference using Gibbs sampling . To our knowledge this is the first attempt at unsupervised preposition sense disambiguation. Our best accuracy reaches 56 a significant improvement at p .001 of 16 over the most-frequent-sense baseline. 1 Introduction Reliable disambiguation of words plays an important role in many NLP applications. Prepositions are ubiquitous they account for more than 10 of the words in the Brown corpus and highly ambiguous. The Preposition Project Litkowski and Hargraves 2005 lists an average of senses for each of the 34 most frequent English prepositions while nouns usually have around two Word-Net nouns average about senses if monose-mous nouns are excluded Fellbaum 1998 . Disambiguating prepositions is thus a challenging and interesting task in itself as exemplified by the Sem-Eval 2007 task Litkowski and Hargraves 2007 and holds promise for NLP applications such as Information Extraction or Machine Given a sentence such as the following In the morning he shopped in Rome we ultimately want to be able to annotate it as 1See Chan et al. 2007 for how using WSD can help MT. 323 in TEMPORAL the morning TIME he PERSON shopped SOCIAL in LOCATIVE Rome LOCATION Here the preposition in has two distinct meanings namely a temporal and a locative one. These meanings are context-dependent. Ultimately we want to disambiguate prepositions not by and for themselves but in the context of sequential semantic labeling. This should also improve disambiguation of .

• Báo cáo khoa học
• Models and Training
• Unsupervised Preposition Sense Disambiguation
• Dirk Hovy and Ashish Vaswani
• báo cáo khoa học
• báo cáo ngôn ngữ
• ngôn ngữ tự nhiên
• Tạp chí khoa học
• Tạp chí Công nghệ
• Quản lý công nghệ
• Science technology models
• Models of linking education
• Training and production business
• Maximum Expected BLEU Training
• Phrase and Lexicon Translation Models
• Xiaodong
• Training Phrase Translation Models
• Leaving One Out
• Joern Wuebker and Arne Mauser and Hermann Ney
• Stochastic Gradient Descent Training
• L1 regularized Log linear Models
• Cumulative Penalty
• xử lý ngôn ngữ tự nhiên
• Beyond Log Linear Models
• Boosted Minimum Error Rate Training
• N best Re ranking
• Models for Sentence Compression
• A Comparison across Domains
• Training Requirements
• Evaluation Measures
• mô hình ngôn ngữ
• Scalable Inference
• Training of Context Rich Syntactic
• Translation Models
• Minimum Risk Annealing
• Training Log Linear Models
• David A
• Smith and Jason Eisner
• Contrastive Estimation
• Unlabeled Data
• Discriminative Training
• Maximum Entropy Models
• Statistical Machine Translation
• Journal of technology and science education
• 8Th international conference on university teaching and innovation
• Flexible training models
• Conference on university teaching and innovation
• University Teaching and Innovaton
• The Universitat Rovira i Virgili
• Management and development
• Vocational training models
• Sustainable multi dimensional poverty reduction
• Center for vocational education
• Local ethnic minority women
• Industrial Revolution 4 0
• Central Highlands region
• Báo cáo sinh học hay
• cách trình bày báo cáo
• báo cáo sinh học
• công trình nghiên cứu sinh học
• tài liệu về sinh học
• Resistance training
• Cardiovascular adaptation
• Animal model
• Aerobic training
• RT induced favorable adaptations
• calorie balance
• metabolic activator
• fitness model
• muscle ratio
• Weight training
• Cardio training secrets
• Khmer minority ethnic women
• Mekong River Delta
• Vocational training
• Vocational change
• Socio economic development
• công nghệ thông tin
• quản trị mạng
• microsoft training and certification
• Overview of Other MSF Models
• Summary of Ph D thesis
• CDIO approach
• Training vocational teachers
• Undergraduate level
• Suggest CDIO based teaching models
• Educational innovation
• International experience
• Teacher education
• Training levels for teachers
• Training models
• cultivation techniques
• Nursery Techniques
• Macadamia tree planting techniques
• Project for Macadamia
• Macadamia fertilization techniques
• Vietnam Agricultural
• North Vietnam
• Raising seedlings
• Macadamia
• grafting
• budding
• cuttings
• obtaining planting stock
• nursery design
• record keeping
• Macadamia Newsletter
• Matacadamia