TAILIEUCHUNG - Báo cáo khoa học: "An Efficient Method for Determining Bilingual Word Classes"

In statistical natural language processing we always face the problem of sparse data. One way to reduce this problem is to group words into equivalence classes which is a standard method in statistical language modeling. In this paper we describe a method to determine bilingual word classes suitable for statistical machine translation. We develop an optimization criterion based on a maximumlikelihood approach and describe a clustering algorithm. We will show that the usage of the bilingual word classes we get can improve statistical machine translation. . | Proceedings of EACL 99 An Efficient Method for Determining Bilingual Word Classes Franz Josef Och Lehrstuhl fur Informatik VI RWTH Aachen - University of Technology Ahornstrafie 55 52056 Aachen Germany och@ Abstract In statistical natural language processing we always face the problem of sparse data. One way to reduce this problem is to group words into equivalence classes which is a standard method in statistical language modeling. In this paper we describe a method to determine bilingual word classes suitable for statistical machine translation. We develop an optimization criterion based on a maximumlikelihood approach and describe a clustering algorithm. We will show that the usage of the bilingual word classes we get can improve statistical machine translation. 1 Introduction Word classes are often used in language modelling to solve the problem of sparse data. Various clustering techniques have been proposed Brown et al. 1992 Jardino and Adda 1993 Martin et al. 1998 which perform automatic word clustering optimizing a maximum-likelihood criterion with iterative clustering algorithms. In the field of statistical machine translation we also face the problem of sparse data. Our aim is to use word classes in statistical machine translation to allow for more robust statistical translation models. A naive approach for doing this would be the use of mono-lingually optimized word classes in source and target language. Unfortunately we can not expect these independently optimized classes to be correspondent. Therefore mono-lingually optimized word classes do not seem to be useful for machine translation see also Fung and Wu 1995 . We define bilingual word clustering as the process of forming corresponding word classes suitable for machine translation purposes for a pair of languages using a parallel training corpus. The described method to determine bilingual word classes is an extension and improvement of the method mentioned in Och and Weber

• Báo cáo khoa học
• An Efficient Method
• Determining Bilingual Word Classes
• Franz Josef Och
• scientific reports
• model language
• process natural language
• Dynamic bit vector frequent itemset
• Frequent weighted utility itemset
• Multibit segment
• Quantitative databases
• Vietnam Journal of Mechanics
• An efficient homogenization method
• The trigonometric interpolation
• The fast fourier transform
• The finite element model
• The microstructure domain
• Data completion for linear symmetric operators cauchy problem
• Linear symmetric operators
• An efficient method via energy like error minimization
• Energy like error minimization
• International Journal of computer science & communication networks
• An efficient multiple keywords search method
• Extended skip graph
• Plurality of keywords
• A bloom filter
• Efficient method for solving broken characters problem
• Recognition of Vietnamese degraded text
• Vietnamese degraded text
• Possible sub graphs
• Simple genomic DNA extraction
• Efficient and universal method
• The extraction of genomic DNA
• Genomic DNA from bacteria
• Microalgae suitable
• PCR based applications
• A green and efficient protocol
• Grinding method
• Four component reaction
• Dihydropyrano [2
• 3 c] pyrazole derivatives
• Synthesized compounds
• Ganoderma lucidum
• Efficient method
• Qualitative screening of ligninolytic enzyme potential
• Qualitative screening
• Highly efficient method
• Oximation of aldehydes
• Structurally diverse aromatic aldehydes
• Corresponding aldoxime derivatives
• Bis thiourea complexe
• Industrial applications
• Automatic recognizing
• Identification card
• Automatic recognizing text fields
• Education research
• Highly efficient pervoskite solar cells
• Simple single step solution method
• Perovskite solar cell
• Wet processing
• Interfacial engineering
• Journal of Automation and Control Engineering
• Efficient transfer learning
• Reinforcement learning using sparse coding
• Normal Q learning
• Means of sparse coding
• Proposed method significantly
• Tạp chí Khoa học
• Ship operation
• Optimum ship speed
• Vietnam Maritime University
• Efficient operation of ships
• bank
• stock
• Stochastic Volatility
• Efficient Method of Moments (EMM)
• Re
• BMC Chemistry
• Bioanalytical method validation
• Cell culture medium
• 5 Hydroxytryptamine
• RIN 14B cells
• Sigmoid function
• Di erential evolution
• Image contrast enhancement method
• Contrast enhancement using the sigmoid function
• Journal of Chemistry
• Transfer hydrogenation
• Ligand free method
• Electrothermal 9200 melting point apparatus
• p chloroacetophenone
• Efficient second order accurate scheme
• Solving two dimensional
• Two dimensional shallow flow model
• Second order accuracy
• Finite Volume Method
• Efficient Multiplication Methods
• Booth’s Multiplication Method
• A review of modular multiplication
• Respective Hardware Implementations
• The Multiplication algorithm
• Karatsuba Ofman Method
• BMC Biotechnology
• S phenyl L cysteine
• Tryptophan synthase
• Protease inhibitor
• Human immunodeficiency virus
• Chemoenzymatic method
• báo cáo khoa học
• báo cáo y học
• công trình nghiên cứu về y học
• tài liệu về y học
• cách trình bày báo cáo