TAILIEUCHUNG - Báo cáo khoa học: "Modeling with Structures in Statistical Machine Translation"

Most statistical machine translation systems employ a word-based alignment model. In this paper we demonstrate that word-based alignment is a major cause of translation errors. We propose a new alignment model based on shallow phrase structures, and the structures can be automatically acquired from parallel corpus. This new model achieved over 10% error reduction for our spoken language translation task. | Modeling with Structures in Statistical Machine Translation Ye-Yi Wang and Alex Waibel School of Computer Science Carnegie Mellon University 5000 Forbes Avenue Pittsburgh PA 15213 USA yyw waibel Abstract Most statistical machine translation systems employ a word-based alignment model. In this paper we demonstrate that word-based alignment is a major cause of translation errors. We propose a new alignment model based on shallow phrase structures and the structures can be automatically acquired from parallel corpus. This new model achieved over 10 error reduction for our spoken language translation task. 1 Introduction Most if not all statistical machine translation systems employ a word-based alignment model Brown et al. 1993 Vogel Ney and Tillman 1996 Wang and Waibel 1997 which treats words in a sentence as independent entities and ignores the structural relationship among them. While this independence assumption works well in speech recognition it poses a major problem in our experiments with spoken language translation between a language pair with very different word orders. In this paper we propose a translation model that employs shallow phrase structures. It has the following advantages over word-based alignment Since the translation model can directly depict phrase reordering in translation it is more accurate for translation between languages with different word phrase orders. The decoder of the translation system can use the phrase information and extend hypothesis by phrases multiple words therefore it can speed up decoding. The paper is organized as follows. In section 2 the problems of word-based alignment models are discussed. To alienate these problems a new alignment model based on shallow phrase structures is introduced in section 3. In section 4 a grammar inference algorithm is presented that can automatically acquire the phrase structures used in the new model. Translation performance is then evaluated in section 5 and conclusions are .

• Báo cáo khoa học
• Modeling with Structures
• Statistical Machine Translation
• Long Papers
• báo cáo khoa học
• mô hình ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Autodesk inventor
• Lecture Autodesk inventor
• Solid modeling data structures
• Solid modeling
• Boundary representation modeling
• B rep data structure
• Mechanical sciences
• Mesh free micromechanical modeling
• Inverse opal structures
• Discrete element method
• Bonded particle method
• Fully coupled computational modeling
• The dynamic response
• 1 3 multiferroic composite structures
• Magnetoelectric coupling
• Service life
• RC structures
• Numerical modeling
• Chloride penetration
• Probabilistic manner
• Protein structure prediction
• Model refinement
• Loop modeling
• Terminus modeling
• Protein structures
• Homologous proteins
• Hybrid structures
• Interface stiffnesses
• Asymptotic algorithm
• Material layers in hybrid structures
• Contact interface between two material layers
• Antenna and EM Modeling with Matlab
• Design of antennas
• Wireless communications
• Cylindrical structures
• Impedance matrix calc jlations
• Antenna circut
• Discrete mathematics
• Discrete mathematics applications
• Discrete Structures
• Mathematical Reasoning
• Combinatorial Analysis
• Modeling Computation
• Wind turbine
• Statistical modeling
• Quantitative evaluation
• Variable speed electronics
• Yaw drive
• Holonomic non linear modelling
• Heterogeneously resisting structures
• Drastic simplifications of the fabric
• Non linear analyses
• Full mathematical modeling
• Modeling of Ring Resonators
• Modes
• Perturbations
• Electronically Tunable
• Electronically Switchable
• Filter Applications
• báo cáo hóa học
• công trình nghiên cứu về hóa học
• tài liệu về hóa học
• cách trình bày báo cáo
• Number theory
• Advanced counting techniques
• Set cardinality
• Language of sets
• Credit Derivatives
• Credit Linked Notes
• Synthetic Collateralized Debt Obligation Structures
• Credit Risk Modeling
• Credit Default Swaps
• Credit Related Spread Products
• BMC Bioinformatics
• Protein protein interactions
• Protein docking
• Structural bioinformatics
• Machine learning
• Gene Ontology filters
• Genome Biology
• Super resolution microscopy
• Polymer models
• Chromatin fiber
• Cell nuclei plays
• Development Methodology
• Trading System
• Control Structures
• Financial Markets
• XML Protocols
• Approach to Endometriosis
• Endometriosis
• Long Term Care
• Geriatric Urinary
• Incontinence – Special
• Frail Elderly
• Physical interactions
• Biological processes
• Homology modeling