臺灣博碩士論文加值系統

英文是一個重要的資訊傳播橋梁，所以要如何從英文文章內汲取有用的資訊就顯得相當重要，一篇文章的重點通常可以由多個關鍵詞表示。如何在英文文章中自動探勘出關鍵詞，並有效地讓使用者明白文章的重點，是近年來一項重要的研究議題。然而，部分關鍵詞萃取方法忽略了文章和關鍵詞間主題資訊的關聯性，因此本論文使用Topic PageRank (TPR)技術萃取全文關鍵詞，該方法基於文章的關鍵詞與該文章的主題相關度來決定一個關鍵詞的權重分數。再者，也創新的將TPR運用於段落的關鍵詞萃取，再結合本論文提出的Google Rank(GR)技術萃取段落關鍵詞，稱為TGPR。最後，產生改良式的三階層概念圖，考慮不同層關鍵詞的共現程度，強化關鍵詞之關聯性，將文章主題有效地呈現。三個資料集的實驗結果顯示，TPR在萃取全文關鍵詞時的績效相對優於其他的比較方法，在萃取段落關鍵詞時的績效亦相對於其他方法優異。

English often plays an important role in information transmission. For this reason, how to extract useful message from English articles is very important. Keyphrases often represent the summary of an article. In recent year, how to mine keyphrases and effective presentation is an important research topic. However, most of the approaches ignored the relationship among the article, the topic and keyphrases. As a result, in this study we integrate the Topic PageRank(TPR) technique to the process of keyphrase extraction of an article. The idea behind TPR is if the topic of the article and a phrase matches together, the phrase shall be viewed as an important phrase for the article. Moreover, TPR is applied innovatively to keyphrase extraction of paragraphs. Finally, TPR and our Google Rank(GR) are further integrated to form a novel Topic Page Google Rank(TGPR) that builds state-of-the-art concept map. Experimental results indicate that the proposed approach outperforms other approaches on keyphrase extraction. Moreover, we propose an improved three-layer concept map which presents the extracted keyphrases to users. The improved map considers the co-occurrence relationship of the keyphrases in the second- and the third-layer so as to enhance consistency of the map.

摘要...........................................................................i
Abstract......................................................................ii
誌謝.........................................................................iii
目錄..........................................................................iv
表目錄.........................................................................v
圖目錄........................................................................vi
一、緒論.......................................................................1
二、文獻探討...................................................................4
2.1 關鍵詞萃取.................................................................4
2.2主題模型....................................................................5
2.3概念圖......................................................................6
三、方法.......................................................................8
3.1系統架構....................................................................8
3.2訓練主題模型................................................................8
3.3 預處理.....................................................................9
3.4 建構字圖..................................................................10
3.5 關鍵詞萃取................................................................12
3.6 建立概念圖................................................................15
四、實驗結果..................................................................19
4.1實驗資料集.................................................................19
4.2評估標準...................................................................20
4.3評估結果...................................................................20
4.4概念圖呈現.................................................................28
五、結論......................................................................30
六、參考文獻..................................................................31

1. Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent Dirichlet Allocation. Journal of Machine Learning Researc, 3, 3/1/2003 993-1022.

2. Bruillard, E., & Baron, G.-L. (2000). Computer-based concept mapping: a review of a cognitive tools for students. Paper presented at the Proceedings of the International Conference on Educational Uses of Communication and Information Technologies, Beijing, China.

3. Chen, J.-R. (2014). Key-Concept Extraction for Concept Map Construction. (Master), National Yunlin University of Science & Technology, Douliu, Yunlin.

4. Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41(6), 391-407.

5. Ercan, G., & Cicekli, I. (2007a). Using lexical chains for keyword extraction. Information Processing & Management, 43(6), 1705-1714.

6. Ercan, G., & Cicekli, I. (2007b). Using lexical chains for keyword extraction. . Information Processing & Management, 43(6), 9.

7. Frank, E., Paynter, G. W., Witten, I. H., Gutwin, C., Nevill-Manning, & G., C. (1999). Domain-specific keyphrase extraction. Paper presented at the In Proceedings of IJCAI.

8. Hofmann, T. (1999). Probabilistic latent semantic indexing. Paper presented at the Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval New York.

9. Hulth, A. (2003). Improved automatic keyword extraction given more linguistic knowledge. Paper presented at the In Proceedings of EMNLP.

10. Jain, G. P., Gurupur, V. P., Jennifer L. Schroeder, & Faulkenberry, E. D. (2014). Artificial Intelligence-Based Student Learning Evaluation: A Concept Map-Based Approach for Analyzing a Student's Understanding of a Topic. IEEE TRANSACTIONS ON LEARNING TECH, 7, 11.

11. Liu, Z., Huang, W., Zheng, Y., & Sun, M. (2010). Automatic Keyphrase Extraction via Topic Decomposition. Paper presented at the Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, MIT,
Massachusetts, USA. retrieved from

12. Mihalcea, R., & Tarau, P. (2004 ). TextRank: Bringing Order into Texts. Paper presented at the In Proceedings of EMNLP.

13. Page, L., Brin, S., Motwani, R., & Winograd, T. (1998). The PageRank Citation Ranking: Bringing Order to the Web: Technical report, Stanford Digital Library Technologies Project.

14. Porter, M. F. (1997). An algorithm for suffix stripping. In J. Karen Sparck & W. Peter (Eds.). Readings in information retrieval, 313-316: Morgan Kaufmann Publishers Inc.

15. Salton, G., & Buckley, C. (1988). Term-weighting approaches in automatic text retrieval. Information processing and management, 24(5), 513–523.

16. Starr, R. R., & Oliveira, J. M. P. d. (2012). Concept maps as the first step in an ontology construction method. Information Systems, 12.

17. Turney, P. D. (2000). Learning algorithms for keyphrase extraction. Information Retrieval, 2(4), 303–336.

18. Wan, X., & Xiao, J. (2008a). CollabRank: Towards a Collaborative Approach to Single-Document Keyphrase Extraction Paper presented at the Computational Linguistics, Manchester.

19. Wan, X., & Xiao, J. (2008b). Single Document Keyphrase Extraction Using Neighborhood Knowledge Paper presented at the Artificial Intelligence American Association.

20. Witten, I. H., Paynter, G. W., Frank, E., Gutwin, C., & Nevill-Manning, C. G. (1999). Kea: Practical automatic keyphrase extraction. Paper presented at the In Proceedings of DL.

21. Yang, Y.-F., Wong, W.-K., & Yeh, H.-C. (2008). A Computer System of Referential Resolution to Assess Students’Reading Comprehension Educational Technology & Society, 11(4), 16.

22. 邱汾柔. (2013). 運用字詞及概念頻率方法於不同結構文章上建構概念圖. (碩士), 雲林科技大學, 雲林縣.

23. 游佳萍, 張玲星, & 黃冠男. (2015). 以概念圖方法探討網路群組學習者的知識建構. 中華民國資訊管理學報, 22(1), 31 - 54.