英語為我國語文教學的一門重要科目。以往的研究顯示，大量的閱讀能增進語文能力，但學生在閱讀後是否理解內容則需要適當的評估方式。文章主角及內容摘要的非選擇問題可瞭解學生是否理解文章內容，但此類型的問答題，若由教師進行評分需花費許多時間，因此本研究將文章摘要問答題進行自動化評分，將可加速評估回饋並增加學生練習的機會。本研究從文章內容擷取特徵，使用機器學習的方法建立模型，進行文章類型自動分類，以挑選合適的文意理解問答題。針對學生回答的摘要結果自動化評分，本研究不需要教師提供答案，而是將英文文章及學生的摘要分別建立語意關係圖，運用語意關係圖計算出各字詞在文章及摘要內容中的重要性，並透過比對英文文章及學生摘要的語意關係圖，取出各種比對特徵，以機器學習的方法建立預測評分等第的分類模型，用來對學生回答的摘要進行語意符合程度自動化評分。實驗結果顯示，本研究所提出的方法在文章有明確的字詞表達文章重點時，可達到不錯的正確率。

English is an important subject of language teaching in our country. Previous studies have shown that a lot of reading can enhance language ability, but it needs appropriate assessment methods to judge that whether students understand the contents after reading. The open questions about article's main role and article's summarization can evaluate whether students understand the content of an article. However, such kind of questions need a lot of time of scoring by teacher. The main goal of this study is to provide automatic scoring for the summarization questions of articles. Accordingly, the students can get evaluation feedback in short time such that it can provide more opportunities for students to practice. In our study, we extract the different features from the content of the article. After that, the machine learning method is used to establish classification model for two article types. According to the article type, suitable questions are selected to be the summarization questions. In the proposed system, teachers are not required to provide answers. Instead, the article and the students' summarizations are represented by semantic graphs in order to calculate the importance score of each word in the article and the students' summarizations, respectively. Then the semantic graphs of the article and the students' summarizations are compared to extract the matching features. Finally, the machine learning method is used to establish the classification model of automatic scoring for the given summarizations. The experiment results show that the proposed method can achieve high accuracy when the articles have distinguishable words to express its focus.

[1] M. Agarwal and P. Mannen, “Automatic Gap-fill Question Generation from Text Books,” in Proceeding IUNLPBEA '11 Proceedings of the 6th Workshop on Innovative Use of NLP for Building Educational Applications, Pages 56-64, 2011.
[2] L. F. Bachman, N. Carr, G. Kamei, M. Kim, M. J. Pan, C. Salvador and Y. Sawaki, “A Reliable Approach to Automatic Assessment of Short Answer Free Responses,” in Proceeding of the 19th international conference on Computational linguistics - Volume 2, Pages 1-4, 2002.
[3] C.-C. Chang and C.-J. Lin, “LIBSVM: A Library for Support Vector Machines,” National Taiwan University, Taipei, Taiwan, 2013.
[4] S. Curto, A. C. Mendes and L. Coheur, “Exploring linguistically-rich patterns for question generation,” in Proceeding of the UCNLG+Eval: Language Generation and Evaluation Workshop, Pages 33-38, 2011.
[5] R. Mitkov and L. A. Ha, “Computer-Aided Generation of Multiple-Choice Tests,” in Proceeding Natural Language Processing and Knowledge Engineering, 2003.
[6] M. Mohler, R. Bunescu and R. Mihalcea, “Learning to Grade Short Answer Questions using Semantic SimilarityMeasures and Dependency Graph,” in Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, Pages 752-762, 2011.
[7] A. J. C. Joseph D. Novak, “The Theory Underlying Concept Maps and How to Construct and Use Them,” Technical Report IHMC CmapTools 2006-01 Rev 01-2008, Florida Institute for Human and Machine Cognition, 2008.
[8] C. P. Ros´e, A. Roque, D. Bhembe and K. VanLehn, “A Hybrid Approach to Content Analysis for Automatic Essay Grading,” in Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology: companion volume of the Proceedings of HLT-NAACL 2003--short papers - Volume 2, Pages 88-90, 2003.

[9] S. G. Pulman and J. Z. Sukkarieh, “Automatic Short Answer Marking,” in Proceedings of the second workshop on Building Educational Applications Using NLP, Pages 9-16, 2005.
[10] E. Sumita, F. Sugaya and S. Yamamoto, “Measuring Non-native Speakers Proficiency of English by Using a Test with Automatically-Generated Fill-in-the-Blank Questions,” in Proceedings of the 2nd Workshop on Building Educational Applications Using NLP, pages 61–68, 2005.
[11] Y.-C. Yang, J.-F. Yang, J.-M. Chang and J. S. Chang, “電腦輔助閱讀測驗自動出題Development of a Computer Assisted Reading Comprehension Test,” in International Conference on English Instruction and Assessment, 2006.
[12] K. Zubrinic, D. Kalpic2 and M. Milicevic, “The automatic creation of concept maps from documents written using morphologically rich languages,” Expert Systems with Applications, page 12709-12718, 2012.
[13] R. Ziai, N. Ott and D. Meurers, “Short Answer Assessment: Establishing Links Between Research Strands, ” In Proceedings of the Seventh Workshop on Building Educational Applications Using NLP, page 190-200, 2012.
[14] 張春興, 教育心理學-三化取向的理論與實踐, 台北市: 東華書局, 1989.
[15] 李秀娟, “不同教學策略對國中生學習生物的影響,” 國立台灣師範大學科學教育研究所碩士論文, 1998.
[16] Stanford Corenlp, http://nlp.stanford.edu/software/corenlp.shtml.
[17] Wordnet, http://wordnet.princeton.edu/.