automatic evaluation of chinese translation output wordlevel or characterlevel
Báo cáo khoa học: "Automatic Evaluation of Chinese Translation Output: Word-Level or Character-Level" doc
... for automatic evaluation of Chinese translation output. Our analysis suggests several key reasons behind this finding. 2 Chinese Translation Evaluation Automatic MT evaluation aims at formulating ... Association for Computational Linguistics Automatic Evaluation of Chinese Translation Output: Word-Level or Character-Level? Maoxi Li Chengqing Zong Hwee Tou Ng National Laboratory of Pattern ... 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...
- 6
- 344
- 1
Tài liệu Báo cáo khoa học: "Automatic Evaluation of Machine Translation Quality Using Longest Common Subsequence and Skip-Bigram Statistics" doc
... human evalua-tors. For the input of automatic evaluation meth-ods, we created three evaluation sets from the MT outputs: 1. Case set: The original system outputs with case information. 2. ... section, we present the evaluations of ROUGE-L, ROUGE-S, and compare their per-formance with other automatic evaluation meas-ures. 5 Evaluations One of the goals of developing automatic evalua-tion ... 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...
- 8
- 442
- 0
Báo cáo khoa học: "a Method for Automatic Evaluation of Machine Translation" pot
... Proceedings of the First Conference of the Association for Machine Translation in the Ameri-cas, pages 193–205, Columbia, Maryland. BLEU: a Method for Automatic Evaluation of Machine Translation Kishore ... judgeswhich substitutes for them when there isneed for quick or frequent evaluations.11 Introduction1.1 RationaleHuman evaluations of machine translation (MT)weigh many aspects of translation, including ... “perfect” translations of agiven source sentence. These translations may varyin word choice or in word order even when they usethe same words. And yet humans can clearly dis-tinguish a good translation...
- 8
- 336
- 0
Tài liệu Báo cáo khoa học: "Automatic Evaluation of Sentence-Level Fluency Andrew Mutton∗" pdf
... Method for Automatic Evaluation of Ma-chine Translation. Technical Report RC22176, IBM.Joseph Turian, Luke Shen, and I. Dan Melamed. 2003. Evalua-tion of Machine Translation and its evaluation. ... types of language model forthe generated text. Then in Section 6 we conclude.2 Related WorkIn terms of human evaluation, there is no uniformview on what constitutes the notion of fluency, or ... letters forcorrect spelling, good grammar, rhythm and flow,appropriateness of tone, and several other specificcharacteristics of good text.In terms of automatic evaluation, we are not aware of any...
- 8
- 507
- 0
Báo cáo khoa học: " A Tool for Error Analysis of Machine Translation Output" doc
... Morpho-syntactic information for au-tomatic error analysis of statistical machine translation output. In Proceedings of WMT, pages 1–6, New YorkCity, New York, USA.Matthew Snover, Bonnie Dorr, ... though the main purpose of BLAST is forerror annotation of machine translation output, thefreedom in the use of error typologies and supportannotations also makes it suitable for other taskswhere ... colored text, where the color depends onthe main type of the error. For each new annotationthe user selects the word or words that are wrong,and selects an error type. In figure 1, the words...
- 6
- 479
- 0
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 ... Agreement: The word order of a source sentence agrees substantially withthat of a target sentence. This measure en-sures that the cost of word order adjustmentis small.ãWord Translation Stability: ... trans./word4.21Rewritten TranslationsOriginal TranslationsTable 1: Comparison of TDMT Training Translations and Original Translationsthe rule writers rewrote translations to make tar-get words as simple...
- 8
- 345
- 0
Báo cáo khoa học: "Automatic Evaluation of Linguistic Quality in Multi-Document Summarization" pptx
... 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 ... pair-wise comparisons of competing systemsover a test set of several inputs and 70%for ranking summaries of a specific input.1 IntroductionEfforts for the development of automatic text sum-marizers ... 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....
- 11
- 407
- 0
Báo cáo khoa học: "Correlating Human and Automatic Evaluation of a German Surface Realiser" doc
... measures of fluency, i.e. syntactic cor-rectness.In this paper, we carry out experiments to corre-late automatic evaluation of the output of a surfacerealisation ranking system for German ... 0.3333Table 2: Correlation between human judgementsfor experiment A (rank 1–3) and automatic metricsfor attempting to account for different but equiva-lent translations of a given source word, typicallyby ... between the automatic eval-uation of system output and human (expert andnon-expert) evaluation of the same data (Englishweather forecasts). Their findings show that theNIST metric correlates...
- 4
- 285
- 0
Báo cáo khoa học: "Feedback Cleaning of Machine Translation Rules Using Automatic Evaluation" pot
... 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 ... domain, bilingual corpora of the new do-main are necessary. However, the sizes of the newcorpora are generally smaller than that of the orig-inal corpus because the collection of bilingual sen-tences ... 149,882Corpus# of Words 868,087 984,197 Evaluation # of Sentences 10,145Corpus# of Words 59,533 67,554Test # of Sentences 10,150Corpus# of Words 59,232 67,193Table 1: Corpus Sizethe rule subsets cannot...
- 8
- 313
- 0
Tài liệu Báo cáo khoa học: "Collecting a Why-question corpus for development and evaluation of an automatic QA-system" pdf
... development of sys-tems. Automatic evaluation requires a cor-pus of questions and answers, a definition of what is a correct answer, and a way to com-pare the correct answers to automatic answersproduced ... selection of natural questions. Thearticles varied in topic, degree of formality and theamount of details; from ”Horror film” and ”Christ-mas worldwide” to ”G-Man (Half-Life)” and ”His-tory of London”. ... andTsunenori Mori. 2007. An Overview of the 4th Ques-tion Answering Challenge (QAC-4) at NTCIR work-shop 6. In Proceedings of the Sixth NTCIR WorkshopMeeting, pages 433–440.Chiori Hori, Takaaki Hori,...
- 9
- 610
- 1
Báo cáo khoa học: "A Formula Finder for the Automatic Synthesis of Translation Algorithms" docx
... ing, etc., from English word forms, thereby re- ducing the inflected word forms to more or less stand- ard stem forms. Machinable rules for the automatic splitting of word affixes, a process ... order according to rule (1), added new English correspondents according to rule (2), deleted the translations of homographic Russian words according to rule (3), inserted short words ac- cording ... English correspondent, the inflection of a cor- respondent into the plural, etc.2 (2) A determiner formula Dr for the desired algor- ithm. This is the portion of the algorithm known beforehand;...
- 14
- 432
- 0
Báo cáo khoa học: "Automatic Adaptation of Annotation Standards: Chinese Word Segmentation and POS Tagging – A Case Study" potx
... 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 ... representation of Ng and Low (2004). For word segmentationonly, there are four boundary tags:ã b: the begin of the wordã m: the middle of the wordã e: the end of the wordã s: a single-character wordwhile ... Processingnorms of modern chinese corpus. Technical report.Yue Zhang and Stephen Clark. 2007. Chinese seg-mentation with a word-based perceptron algorithm.In Proceedings of the 45th Annual Meeting of...
- 9
- 404
- 0
Báo cáo khoa học: "Automatic Evaluation Method for Machine Translation using Noun-Phrase Chunking" pptx
... Utsuro and Noriko Kando.2009. Meta -Evaluation of Automatic Evaluation Methods for Machine Translation using Patent Translation Data in NTCIR-7. In Proc. of the3rd Workshop on Patent Translation, ... Na-tional Institute of Informatics (NII) providedcorpora used in this work. The author grate-fully acknowledges JAPIO and NII for theirsupport. Moreover, this work was partiallysupported by Grants ... a parameter for theweight of scorenp: it is between 0.0 and 1.0.The ratio of scorewdto scorenpis 1:1 when δ is1.0. Moreover, scorewd multiand scorenp multiare used for Eq. (13) in...
- 10
- 415
- 0
Báo cáo khoa học: "Towards Web-Based Evaluation of Automatic Natural Language Phrase Generation" potx
... the length, i.e., number of words, among theanalyzed phrases differs and our algorithm doesnot use semantical information, in order to avoidthe distortion of the original phrase, in our systemthe ... generator for open do-main dialogue systems, or chatbots.In this paper, we propose the automatic gene-ration of trivial dialogue phrases through the ap-plication of a transfer-like genetic algorithm ... which one of the genera-ted expressions is correct. Our method differs inthe application of a GA-like transfer process inorder to automatically insert new features on theselected original...
- 4
- 260
- 0
Báo cáo khoa học: "Comparing Automatic and Human Evaluation of NLG Systems" potx
... score for a set of generator forecastsis the average of scores for individual forecasts.Human evaluations: We recruited 9 experts(people with experience reading forecasts for off-shore oil ... types of NLG systems for whichthe correlation of automatic and human evaluation was particularly good or bad.Data: We extracted from each forecast in theSUMTIME corpus the first description of ... top of Table 1).We randomly selected a subset of 21 forecastdates for use in human evaluations. For these 21forecast dates, we also asked two meteorologistswho had not contributed to the original...
- 8
- 376
- 0
Tìm thêm: xác định các mục tiêu của chương trình xác định các nguyên tắc biên soạn khảo sát các chuẩn giảng dạy tiếng nhật từ góc độ lí thuyết và thực tiễn khảo sát chương trình đào tạo của các đơn vị đào tạo tại nhật bản điều tra với đối tượng sinh viên học tiếng nhật không chuyên ngữ1 khảo sát thực tế giảng dạy tiếng nhật không chuyên ngữ tại việt nam khảo sát các chương trình đào tạo theo những bộ giáo trình tiêu biểu nội dung cụ thể cho từng kĩ năng ở từng cấp độ xác định mức độ đáp ứng về văn hoá và chuyên môn trong ct phát huy những thành tựu công nghệ mới nhất được áp dụng vào công tác dạy và học ngoại ngữ mở máy động cơ lồng sóc mở máy động cơ rôto dây quấn hệ số công suất cosp fi p2 đặc tuyến hiệu suất h fi p2 đặc tuyến mômen quay m fi p2 đặc tuyến tốc độ rôto n fi p2 đặc tuyến dòng điện stato i1 fi p2 sự cần thiết phải đầu tư xây dựng nhà máy phần 3 giới thiệu nguyên liệu từ bảng 3 1 ta thấy ngoài hai thành phần chủ yếu và chiếm tỷ lệ cao nhất là tinh bột và cacbonhydrat trong hạt gạo tẻ còn chứa đường cellulose hemicellulose