5W1H Training Effectiveness for Information Extraction: Interpreting Summarized Chinese Indictments into English

檢視運用5W1H資料擷取策略於口譯訓練之成效：以簡易起訴書中譯英為例

10.29912/CTR.202003_13(1).0004

張中倩(Karen Chung-chien Chang)

5W1H strategy ； information extraction ； summarized indictment ； indictment interpreting ； 5W1H策略 ； 訊息擷取 ； 簡易起訴書 ； 起訴書翻譯

編譯論叢

13卷1期（2020 / 03 / 01）

115 - 164

英文

In the past decade, court translation/interpretation has attracted much research interest in Taiwan. In 2008, relevant guidelines and details were established, and court translation and interpretation became formal practices. However, some problems have become more noticeable and have drawn the attention of the authorities and researchers (Chang, 2013, 2016; Y. L. Chen, 2018; Chen & Chen, 2013; Y. T. Chen, 2018; Tu, 2019). One problematic issue is the lack of training for certified court interpreters. Because the threshold of language competence for becoming a court-certified interpreter is not very high, the assumption that a certified interpreter can effectively assist with a court case has received much criticism. This difficulty stems from the fact that most legal documents are not reader-friendly, a feature further compounding the comprehension problem when all messages are conveyed orally. As the syntactic structures of Chinese and English differ substantially, novice court interpreters, when hearing a paragraph of condensed expressions, tend to become baffled and unsure of where to start. Consequently, much time is required for information processing. This situation is especially evident in court interpreters' handling of summarized indictments. The present study adopts a training method to help student interpreters to parse a summarized indictment and extract key information. Employing the 5W1H strategy, this study examines the effect of training on students' ability to process the given indictments in Chinese and to interpret them accurately into English. The participants processed 13 indictments in total, and their renditions were evaluated based on the criteria of information completeness and grammatical correctness. The results have indicated the 5W1H strategy greatly helped the participants filter through the layers of information more effectively and produce English interpreting renditions more accurately.

過去10年，法庭口譯成為新的研究議題。於2008年開始，臺灣設立、實施法庭通譯之相關規定及細節，但有關當局、學界均注意到一些主要問題（Chang, 2013, 2016; Y. L. Chen, 2018; Chen & Chen, 2013; Y. T. Chen, 2018; Tu, 2019）。其中，法庭通譯訓練的缺乏為主要議題之一，由於成為法庭通譯的語言能力門檻不高，在特約通譯取得證書後，是否有能力協助處理法庭案件，儼然引起批評及憂慮。由於多數法律文件內容艱澀，在法庭訊問中，以閱讀方式表達時，更容易造成通譯人員在聆聽訊息後，無法迅速處理訊息及翻譯；再加上中、英語言句法結構上存在許多差異，對於新手法庭通譯來說，當聽到一大段資料濃縮的訊息時，常常不知如何著手進行翻譯。在案件處理時，此情況常發生於簡易起訴書的翻譯。本研究採用5W1H資料擷取策略訓練學習「法庭口譯」的學生，檢視此策略是否可以有效地幫助學習者擷取簡易起訴書中的重要訊息，在檢視學習成效時，重點為中譯英的「信息完整度」及「文法正確性」，學生共處理了13件簡易起訴書，分析顯示：5W1H策略對於訓練訊息擷取非常有成效，學習者能擷取、處理案件所提之多重細節（人物、地點、內容、時間、緣由、過程），在13週訓練之後，學生在斷句、擷取細節、中譯英的表達及文法正確度方面，皆有明顯進步。

1. Chang, K. C. C.(2016).Needs analysis for the training of court interpreters.Compilation and Translation Review,9(2),93-136.
   連結：
2. Chang, K. C. C.(2013).Current practices of court interpreting in Taiwan: Challenges and possible solutions.Compilation and Translation Review,6(2),127-164.
   連結：
3. Chen, Y. L.,Chen, T.(2013).Enhancing the quality of court interpretation – A functionalist approach.Compilation and Translation Review,6(2),99-126.
   連結：
4. Chen, Y. L.,Liao, P.(2016).A revised model for the professionalization of court interpreting in Taiwan.Compilation and Translation Review,9(2),137-164.
   連結：
5. Alvarez, R.,Vidal, M. C.(1996).Translation, power, subversion.Philadelphia, PA:Multilingual Matters.
6. Berk-Seligson, S.(1999).The impact of court interpreting on the coerciveness of leading questions.The Journal of Forensic Linguistics,6,30-56.
7. Carmagnola, F.(2008).The five Ws in user model interoperability.5th International Workshop on Ubiquitous User Modeling,Gran Canaria, Spain:
8. Chakma, K.,Das, A.(2018).A 5W1H based annotation scheme for semantic role labeling of English tweets.Computacióny Sistemas,22(3),747-755.
9. Chen, Y. L.(2018).Court interpreting: Theory and practice.Taipei, Taiwan:Wu-Nan Book.
10. Chen, Y. T.(2018).Taiwan,National Taiwan Normal University.
11. Chinchor, N.,Marsh, E.(1998).MUC-7 information extraction task definition (version 5.1).Seventh Message Understanding Conference (MUC-7)
12. Cowie, J.,Wilkes, Y.(2000).Information extraction.Handbook of natural language processing,New York, NY:
13. Ding, R.,Li, Z.(2018).Event extraction with deep contextualized word representation and multi-attention layer.14th International Conference: Advanced Data Mining and Applications (ADMA 2018),Cham, Switzerland:
14. González, R. D.,Vásquez, V. F.,Mikkelson, H.(1991).Fundamentals of court interpretation: Theory, policy and practice.Durham, NC:Carolina Academic Press.
15. Hale, S.(1999).Interpreters’ treatment of discourse markers in courtroom questions.The Journal of Forensic Linguistics,6(1),57-82.
16. Hale, S.(2004).The discourse of court interpreting.Amsterdam, the Netherlands:John Benjamins.
17. Han, S.,Kim, Y.,Lee, G. G.(2015).Micro-counseling dialog system based on semantic content.Natural language dialog systems and intelligent assistants,Cham, Switzerland:
18. Han, S.,Lee, K.,Lee, D.,Lee, G. G.(2013).Counseling dialog system with 5W1H extraction.Proceedings of the SIGDIAL 2013 Conference,Stroudsburg, PA:
19. Hayes, A. F.,Krippendorff, K.(2007).Answering the call for a standard reliability measure for coding data.Communication Methods and Measures,1,77-89.
20. Hogenboom, F. P.,Frasincar, F.,Kaymak, U.,de Jong, F. M. G.(2011).An overview of event extraction from text.Proceedings of Detection, Representation, and Exploitation of Events in the Semantic Web (DeRiVE 2011), workshop in conjunction with the 10th International Semantic Web Conference 2011 (ISWC 2011),Aachen, Germany:
21. Ikeda, T.,Okumura, A.,Muraki, K.(1998).Information classification and navigation based on 5W1H of the target information.36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics
22. Li, K. X.,Zhang, X. H.(2006).Legal texts and legal translation.Beijing, China:China Translation and Publishing Corporation.
23. Mason, M.(2008).Courtroom interpreting.New York, NY:University Press of America.
24. Morris, R.(1999).The gum syndrome: Predicaments in court interpreting.The Journal of Forensic Linguistics,6(1),7-29.
25. Sarawagi, S.(2007).Information extraction.Foundations and Trends in Databases,1(3),261-377.
26. Sharma, S.,Kumar, R.,Bhadana, P.,Gupta, S.(2013).News event extraction using 5W1H approach & its analysis.International Journal of Scientific & Engineering Research,4(5),2064-2067.
27. Shimazu, K.,Arisawa, T.,Saito, I.(2006).Interdisciplinary contents management using 5W1H interface for metadata.2006 IEEE/WIC/ACM International Conference on Web Intelligence (WI2006 main conference proceedings) (WI’06),Los Angeles, CA:
28. Tu, H. L.(2019).Taiwan,National Taiwan Normal University.
29. Wang, W.(2012).Chinese news event 5W1H semantic elements extraction for event ontology population.WWW’12-Proceedings of the 21st Annual Conference on World Wide Web Companion,New York, NY:
30. Wang, W.,Zhao, D.(2012).Ontology-based event modeling for semantic understanding of Chinese news story.NLPCC 2012, CCIS 333,Heidelberg, Germany:
31. Wang, W.,Zhao, D.,Zou, L.,Wang, D.,Zheng, W.(2010).Extracting 5W1H event semantic elements for Chinese online news.Web-Age Information Management: 11th International Conference, WAIM 2010 proceedings,Berlin, Germany:
32. Yule, G.(1996).Pragmatics.Oxford, UK:Oxford University Press.
33. Zheng, L.,Jin, P.,Zhao, J.,Yue, L.(2014).A fine-grained approach for extracting events on Microblogs.Database and Expert Systems Applications, 25th International Conference, DEXA 2014, Munich, Germany, September 1-4, 2014. Proceedings, part I,Cham, Switzerland: