Báo cáo khoa học: "Automatic Evaluation of Chinese Translation Output: Word-Level or Character-Level" doc

... LinguisticsAutomatic Evaluation of Chinese Translation Output: Word-Level or Character-Level? Maoxi Li Chengqing Zong Hwee Tou Ng National Laboratory of Pattern Recognition Institute of Automation, Chinese ... translation segmented into words: Translation: 多少_钱_的_伞_吗_？ Reference: 这些_雨伞_多少_钱_？ The word “伞” is a synonym for the word “雨伞”, and both words are translations of the English word ... character-level metrics more suitable for the automatic evaluation of Chinese translation output. 6 Conclusion In this paper, we conducted a detailed study of the relative merits of word-level versus...

Tài liệu Báo cáo khoa học: "Automatic Evaluation of Machine Translation Quality Using Longest Common Subsequence and Skip-Bigram Statistics" doc

... construction of N-best translation lexicons from parallel text. Melamed (1995) used the ratio (LCSR) between the length of the LCS of two words and the length of the longer word of the two words ... auto-matic evaluation measure. To apply LCS in machine translation evaluation, we view a translation as a sequence of words. The intuition is that the longer the LCS of two transla-tions is, the more ... recall reflects the proportion of words in X (refer-ence translation) that are also present in Y (candi-date translation) ; while unigram precision is the proportion of words in Y that are also...

Tài liệu Báo cáo khoa học: "Automatic Evaluation of Sentence-Level Fluency Andrew Mutton∗" pdf

... to the main purpose of the SVM, using distancefrom support vector as a metric. Results are givenfor correlation of GLEU against sequence sizes forall data (Table 2) and for the human trial ... ‘moderate’ range for correlation in-terpretation. In particular, for the GLEU–human cor-relation, the score of 0.4014 is approaching the min-imum pairwise human correlation of 0.4710.5 Different ... returns fragments for Everybody likes big cakesand for do. We expect that the number of fragmentsshould correlate inversely with the quality of a sen-tence. For a metric, we normalise this number...

Báo cáo khoa học: "Automatic Evaluation of Linguistic Quality in Multi-Document Summarization" pptx

... of con-tent selection, which allows for frequent evalua-tion during system development and for report-ing results of experiments performed outside of the annual NIST-led evaluations, the DocumentUnderstanding ... of systems,ignoring the linguistic quality of the system out-put. Part of the reason for this imbalance is theexistence of ROUGE (Lin and Hovy, 2003; Lin,2004), the system for automatic evaluation ... are the total number of word types fromboth sentences siand si+1. Stop words were re-tained. The value of each dimension for a sentenceis the number of tokens of that word type in thatsentence....

Báo cáo khoa học: "Automatic Construction of Machine Translation Knowledge Using Translation Literalness" docx

... LanguageMean # of Translation Words per Source WordMean Context-freeness (# of Word Link = 4)28,300 words (49.5%)3,107 words1.51 trans./word4.4520,722 words (34.0%)3,601 words1.94 trans./word4.21Rewritten ... For example, word-level statistical MT (Brown et al., 1993) translates asource sentence with a combination of word trans-fer and word order adjustment. Thus, word-order agreement is an important ... not. About 13% of different Englishsentences have multiple Japanese translations. Translation Dictionary: Extraction of WordCorrespondence For word correspondencesthat occur more than nine...

Báo cáo khoa học: "AUTOMATIC ACQUISITION OF A LARGE SUBCATEGORIZATION DICTIONARY FROM CORPORA" doc

... training corpus). PREVIOUS WORK While work has been done on various sorts of col- location information that can be obtained from text corpora, the only research that I am aware of that has ... ACQUISITION OF A LARGE SUBCATEGORIZATION DICTIONARY FROM CORPORA Christopher D. Manning Xerox PARC and Stanford University Stanford University Dept. of Linguistics, Bldg. 100 Stanford, CA 94305-2150, ... conciliatory terms. But the majority of occurrences of in after a verb are NP modifiers or non-subcategorized locative phrases, such as those in (4). s (4) a. He gauged support for a change...

Báo cáo khoa học: "Automatic Acquisition of Script Knowledge from a Text Collection" docx

... paragraphs in clusters based onthe date of issue of the report. We used only thefirst paragraphs of the news reports because theytend to describe facts in time order.After that, the sentences in ... objects. A 'pair of actions'consists of two actions that occur in time order. A'sequence of actions' can be defined as a transitiveclosure of all the pairs of actions.1. Cases ... for automatic acquisition of script knowl-edge and investigated the effectiveness of ourmethod. We used issues of Nihon Keizai Shim-bun for the past 11 years (1990-2000) as a news-paper corpus...

Báo cáo khoa học: "Feedback Cleaning of Machine Translation Rules Using Automatic Evaluation" pot

... ScoreNumber of RulesNumber of IterationsTest Corpus BLEU Score Evaluation Corpus BLEU ScoreNumber of RulesFigure 5: Relationship between Number of Itera-tions and BLEU Scores/Number of ... Cross-cleaningIn general, most evaluation corpora are smaller thantraining corpora. Therefore, omissions of cleaningTrainingCorpusTraining Evaluation Training Evaluation Training Evaluation TrainingBaseRule ... size of the evaluation corpus. Furthermore, inorder to ﬁnd the nearest neighbor solution, we haveto calculate all BLEU scores of the neighborhood.If the number of rules is R and the neighborhoodis...

Báo cáo khoa học: "Automatic Adaptation of Annotation Standards: Chinese Word Segmentation and POS Tagging – A Case Study" potx

... representation of Ng and Low (2004). For word segmentationonly, there are four boundary tags:• b: the begin of the word• m: the middle of the word• e: the end of the word• s: a single-character wordwhile ... point for wordsegmentation and 1 point for Joint S&T, with cor-responding error reductions of 30.2% and 14%.The final result outperforms the latest work on thesame corpus which uses more ... first report experimental results of the singleperceptron classifier on CTB 5.0. The originalcorpus is split according to former works: chap-ters 271 − 300 for testing, chapters 301 − 325 fordevelopment,...

Tài liệu Báo cáo khoa học: "Automatic Extraction of Lexico-Syntactic Patterns for Detection of Negation and Speculation Scopes" pdf

... next performed a simple transfor-mation of the derived rule set. If all children of arule tree node are of type *scope* or * (i.e. non-cue words), the node label is replaced by *scope* or * respectively, ... tree; neighboring identical siblings of type *scope* or * are replaced by a single node of the corresponding type. Figure 3 shows an example of this transformation.(a) The children of nodes JJ/NN/NN ... Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics:shortpapers, pages 283–287,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational...

Xem thêm