Discriminative Models for Automatic Acquisition of Translation Equivalences

Chun-Xiang Zhang,Sheng Li,Tie-Jun Zhao 대한전기학회 2007 International Journal of Control, Automation, and Vol.5 No.1

Translation equivalence is very important for bilingual lexicography, machine translation system and cross-lingual information retrieval. Extraction of equivalences from bilingual sentence pairs belongs to data mining problem. In this paper, discriminative learning methods are employed to filter translation equivalences. Discriminative features including translation literality, phrase alignment probability, and phrase length ratio are used to evaluate equivalences. 1000 equivalences randomly selected are filtered and then evaluated. Experimental results indicate that its precision is 87.8% and recall is 89.8% for support vector machine.

Disambiguate Chinese Word Sense Based on Linguistics Knowledge

Chun-Xiang Zhang,Long Deng,Xue-Yao Gao,Zhi-Mao Lu 보안공학연구지원센터 2014 International Journal of Database Theory and Appli Vol.7 No.6

Word sense disambiguation (WSD) is important to many application problems in natural language processing fields, such as machine translation, parsing analysis and information retrieval. In this paper, we propose a new method to determine correct sense categories of Chinese words based on linguistics knowledge. The left word string and the right word string around the ambiguous word are respectively analyzed. Their syntactic structures are obtained for determining its intended sense. Syntactic category and part of speech are extracted as disambiguation features. A naive bayesian model is used as the classifier. Experimental results showed that the accuracy rate of classification arrives at 64%. The performance of disambiguation is improved.

Chinese Word Sense Disambiguation Based on Hidden Markov Model

Zhang Chun-Xiang,Sun Yan-Chen,Gao Xue-Yao,Lu Zhi-Mao 보안공학연구지원센터 2015 International Journal of Database Theory and Appli Vol.8 No.6

Word sense disambiguation (WSD) is important for natural language processing. It plays important roles in information retrieval, machine translation, text categorization and topic tracking. In this paper, the transition among senses of words is considered. For an ambiguous word, its semantic codes and its left word’s semantic codes are taken as disambiguation features. At the same time, a new method based on hidden Markov model (HMM) is proposed for Chinese word sense disambiguation. Chinese Tongyici Cilin is used to determine semantic codes of words. HMM is optimized in training corpus. The WSD classifiers based on HMM is tested. Experimental results show that the accuracy of word sense disambiguation is improved.

Chinese Word Sense Disambiguation Based on Beam Search

Zhang Chun-Xiang,He Shan,Gao Xue-Yao,Lu Zhi-Mao 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.11

Research on word sense disambiguation (WSD) is of great importance in natural language processing. In this paper, a new method based on beam search algorithm for Chinese WSD is proposed. By mining potential knowledge between phrase and semantic category in a sentence, this approach can construct its semantic network. It searches an optimal semantic category sequence from a Chinese sentence's semantic network with beam search algorithm, so that correct meanings of ambiguous words can be found from the optimal sequence. Experiments show that a better WSD performance is gotten.

Word Sense Disambiguation Based on Perceptron Model

Zhang Chun-Xiang,Gao Xue-Yao,Lu Zhi-Mao 보안공학연구지원센터 2016 International Journal of Database Theory and Appli Vol.9 No.5

Word sense disambiguation (WSD) is an important research topic in natural language processing field, which is very useful for machine translation and information retrieval. In this paper, a linear combination model based on multiple discriminative features is proposed to determine correct sense of an ambiguous word, in which morphology and part of speech in left and right words around ambiguous word are used as features. Then, perceptron algorithm is applied to optimize the WSD model. Experiments show that the WSD performance is improved after the proposed method is applied.

Automatic Acquisition Of Translation Equivalences From Bilingual Corpus

Chun?xiang Zhang,Tie?jun Zhao,Sheng Li 한국어정보학회 2006 한국어정보학 Vol.8 No.1

Translation equivalence is very useful for bilingual lexicography, machine translation system, cross‐lingual information retrieval and many applications in natural language processing. A linear combination model of multiple features is used to filter extracted equivalences in this paper. Experimental results indicate that performance of the combination model surpasses other classifiers’ in open test. 1000 equivalences labeled by linear combination model are randomly selected and then evaluated. Its F1 measure achieves 88.13%. Its performances surpass those classifiers.

Determine Word Sense Based on Semantic and Syntax Information

Zhang Chun-Xiang,Sun Lu-Rong,Gao Xue-Yao 보안공학연구지원센터 2016 International Journal of Database Theory and Appli Vol.9 No.2

Word sense disambiguation (WSD) plays an important role in natural language processing fields. Semantic category is semantic knowledge and part-of-speech is syntax knowledge. In this paper, word window is opened to get semantic category and part-of-speech of left and right adjacent words around an ambiguous word. A new approach of determining true meanings of ambiguous words based on support vector machine (SVM) is given. The training corpus in SemEval-2007: Task#5 is applied to optimize SVM and the optimized SVM is tested. Experimental results show that the performance of the proposed method is improved.

Comparative Analysis of Water Stress-Responsive Transcriptomes in Drought-Susceptible and -Tolerant Wheat (Triticum aestivum L.)

Yong Chun Li,Fan Rong Meng,Chun Yan Zhang,Ning Zhang,Ming Shan Sun,Jiang Ping Ren,Hong Bin Niu,Xiang Wang,Jun Yin 한국식물학회 2012 Journal of Plant Biology Vol.55 No.5

To understand better the mechanisms that regulate the water stress response in wheat, we conducted a comparative analysis of transcript profiles in roots from two wheat genotypes -- drought-tolerant ‘Luohan No. 2’ (LH)and drought-susceptible ‘Chinese Spring’ (CS). In LH roots,3831 transcripts displayed changes in expression of at least two-fold over the well-watered control when drought treatment was applied. Of these, 1593 were induced while 2238 were repressed. Relatively fewer transcripts were drought-responsive in CS; i.e., 1404 transcripts were induced and 1493 were repressed. In common between LH and CS,569 transcripts were induced and 424 transcripts were repressed. In all, 689 transcripts (757 probe sets) identified from LH and 537 transcripts (575 probe sets) from CS were annotated and classified into 10 functional categories. Among those annotated transcripts from LH and CS that had fold-change ratios of at least 4, 92 induced transcripts were common to both, while 23 transcripts were specifically induced in LH. Gene ontology analysis of these induced genes showed highly significant enrichment for multiple terms related to abiotic stimuli, organic acid biosynthesis,and lipid metabolism. This suggests that these gene groups play important roles during the stress response in LH and CS, and might also be responsible for differences in drought tolerance between those genotypes.

Tumor Necrosis Factor-α Gene Polymorphisms and Risk of Oral Cancer: Evidence from a Meta-analysis

Chen, Fang-Chun,Zhang, Fan,Zhang, Zhi-Jiao,Meng, Si-Ying,Wang, Yang,Xiang, Xue-Rong,Wang, Chun,Tang, Yu-Ying Asian Pacific Journal of Cancer Prevention 2013 Asian Pacific journal of cancer prevention Vol.14 No.12

Numerous studies have been conducted regarding association between TNF-${\alpha}$ and oral cancer risk, but the results remain controversial. The present meta-analysis is performed to acquire a more precise estimation of relationships. Databases of Pubmed, the Cochrane library and the China National Knowledge Internet (CNKI) were retrieved until August 10, 2013. Pooled odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated with fixed- or random-effect models. The heterogeneity assumption was assessed by I-squared test. Among the eight included case-control studies, all were focused on TNF-${\alpha}$-308G>A and four also concerned the TNF-${\alpha}$-238G>A polymorphism. It was found that oral cancer risk were significant decreased with the TNF-${\alpha}$-308G>A polymorphism in the additive genetic model (GG vs. AA, OR=0.19, 95% CI: [0.04, 1.00], P=0.05, I2=68.9%) and the dominant genetic model (GG+GA vs. AA, OR=0.22, 95% CI: [0.06, 0.82], P=0.03, I2=52.4%); however, no significant association was observed in allele contrast (G vs. A, OR=0.70, 95% CI: [0.23, 2.16], P=0.54, I2=95.9%) and recessive genetic models (GG vs. GA+AA, OR=0.72, 95% CI: [0.33, 1.57], P=0.41, I2=93.1%). For the TNF-${\alpha}$-238G>A polymorphism, significant associations with oral cancer risk were found in the allele contrast (G vs. A, OR=2.75, 95% CI: [1.25, 6.04], P=0.01, I2=0.0%) and recessive genetic models (GG vs. GA+AA, OR=2.23, 95%CI: [1.18, 4.23], P=0.01, I2=0.0%). Conclusively, this meta-analysis indicates that TNF-${\alpha}$ polymorphisms may contribute to the risk of oral cancer. Allele G and the GG+GA genotype of TNF-${\alpha}$-308G>A may decrease the risk of oral cancer, while allele G and the GG genotype of TNF-${\alpha}$-238G>A may cause an increase.

2D Geometric Constraint Optimum Solving Based on Problem Decomposition

Xue-Yao Gao,Chun-Xiang Zhang,Zhi-Mao Lu 보안공학연구지원센터 2015 International Journal of Hybrid Information Techno Vol.8 No.1

Constraint solving is widely applied to many fields including computer aided design, 2 dimension (2D) model design and computer aided manufacturing. Geometric constraint solution is a difficult problem because there are a large number of entities and related parameters in 2D sketches. In this paper, a new method which decomposes geometric constraint relations based on entity-parameter graphs is proposed for reducing the size of constraint solution. A geometric constraint problem is decomposed into many independent sub-problems. Then, particle swarm optimization algorithm is used to solve constraint equations in each sub-problem. Solutions of all sub-problems are integrated to obtain the original problem’s solution. In experiments, the proposed method is applied to HUST-CAID system. Experimental results show that the method can effectively solve 2 dimension geometric constraints.