TAILIEUCHUNG - Báo cáo khoa học: "Learning to Predict Case Markers in Japanese"

Japanese case markers, which indicate the grammatical relation of the complement NP to the predicate, often pose challenges to the generation of Japanese text, be it done by a foreign language learner, or by a machine translation (MT) system. In this paper, we describe the task of predicting Japanese case markers and propose machine learning methods for solving it in two settings: (i) monolingual, when given information only from the Japanese sentence; and (ii) bilingual, when also given information from a corresponding English source sentence in an MT context. . | Learning to Predict Case Markers in Japanese Hisami Suzuki Kristina Toutanova1 Microsoft Research One Microsoft Way Redmond WA 98052 USA hi samis kri stout @ Abstract Japanese case markers which indicate the grammatical relation of the complement NP to the predicate often pose challenges to the generation of Japanese text be it done by a foreign language learner or by a machine translation MT system. In this paper we describe the task of predicting Japanese case markers and propose machine learning methods for solving it in two settings i monolingual when given information only from the Japanese sentence and ii bilingual when also given information from a corresponding English source sentence in an MT context. We formulate the task after the well-studied task of English semantic role labelling and explore features from a syntactic dependency structure of the sentence. For the monolingual task we evaluated our models on the Kyoto Corpus and achieved over 84 accuracy in assigning correct case markers for each phrase. For the bilingual task we achieved an accuracy of 92 per phrase using a bilingual dataset from a technical domain. We show that in both settings features that exploit dependency information whether derived from gold-standard annotations or automatically assigned contribute significantly to the prediction of case markers. 1 Introduction why predict case Generation of grammatical elements such as inflectional endings and case markers has become an important component technology particularly in the context of machine translation MT . In an English-to-Japanese MT system for example Japanese case markers which indicate grammatical relations . subject object location of the complement noun phrase to the predicate are among the most difficult to generate appropriately. This is because the case markers often do not correspond to any word in the source language as many grammatical relations are expressed via word order in English. It is also .

• Báo cáo khoa học
• Learning to Predict
• Case Markers in Japanese
• Hisami Suzuki Kristina Toutanova
• báo cáo khoa học
• mô hình ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Ebook Predict wall street
• Predict wall street
• IQ brainteasers
• English vocabulary
• English books
• English learning method for children
• Learning to predict pitch accents
• prosodic boundaries in Dutch
• Erwin Marsi
• Learning Features
• Predict Cue Usage
• Barbara
• scientific reports
• model language
• process natural language
• Genome Biology
• Machine learning
• Machine learning models
• Genomic activity
• Training cell types
• High throughput
• Eukaryotic pathogens
• Predict protein
• Discovering subunit vaccines
• Machine learning algorithms
• 3D genome
• Non coding mutation
• Deep learning
• Whole genome sequences
• Plant pests
• Diseases prediction
• Smart agriculture
• Product green agriculture
• systems constantly
• ordered systems
• constrain one another
• forecast or predict
• complex phenomena
• behavior of birds
• BMC Bioinformatics
• Hypothetical proteins
• Classification features
• Functional genomics
• Cerebral palsy
• Epigenetic biomarkers
• DNA methylation
• Computational statistics
• Predict succinylation sites
• Multiple features
• Grey pseudo amino acid composition
• Information gain
• Ensemble learning algorithm
• Tri disulfide peptide toxins
• Sequential tri disulfide peptides
• Nonknotted STPs
• Insecticidal peptides
• Active learning
• Drug target prediction
• Matrix factorization
• Regression model
• Protein solubility
• Protein solubility prediction
• In silico prediction
• Recombinant protein expression
• Escherichia coli
• BMC Pulmonary Medicine
• Chronic obstructive pulmonary disease
• Primary care
• FEV1/FVC ratio
• BMC Genomics
• Histone modification
• Topologically associated domains
• Sequence information
• BMC Veterinary Research
• Animal shelter
• High euthanization rates
• Prediction models
• Goal programming approach
• Decision support tool
• BMC Cancer
• Accurate prognosis assessment
• Intrahepatic cholangiocarcinoma
• Gradient boosting machine
• Prostate cancer
• Artificial intelligence
• Gene fusion
• Whole slide imaging
• Pathogen host protein interaction networks
• Pathogen host coevolution
• Signaling pathways
• Transfer learning
• BMC Anesthesiology
• In hospital mortality risk
• Temporal convolution network
• Attention mechanism
• Time series