臺灣博碩士論文加值系統

在現代科技的浪潮下，人類的生活有了許多重大的革新，網際網路的發展更帶動了更迅速的資訊傳遞，以學習方面來說，新的學習方式正在逐步的改變傳統學習的習慣，並提供學習者更多元的學習媒介。在中小學的教育體制，問題式學習是一種有效的學習方式，學生可以透過問題確認自我的學習狀態並理解問題所表達的知識與概念。而為了確切提供學習者所需的資料，好的學習媒介管理與分類就成為一個重要的工作，將問題按照其所涵蓋的知識點進行分類，使得使用者取得適合的題目可以更佳的便利，進而達成更好的學習效率。
本篇論文延續中小學學習教材的分類系統，除了針對學習教材規劃資料庫，並提出跨章節的分類系統。傳統的學習方式經常是針對每個不同的單一知識點進行學習，在原本的系統針對此類的問題已有不錯的分類效果，但在大型入學考試與進階的問題中，有些問題同時具備不同的章節的概念，因此我們延續原先的卷積神經網路(Convolutional Neural Network)與支持向量機器(Support Vector Machine)的分類器，設計了針對跨章節問題的分類架構，並與原本分類系統提出的跨章節分類策略進行比較，證明新提出的架構能達到更佳的分類效能。

In the tide of modern technology, there are many significant innovations in human life. The development of the Internet has led to the more rapid delivery of information. From the learning side, the new learning style is gradually changing the habit of traditional learning. In the K-12 system, the question-driven learning is an effective way of learning. The students can confirm their learning status through question exercises and understand the knowledge and concepts expressed by the problem. In order to provide the learning information for the learners, a good learning material management and classification has become an important task. To classify the question according to the knowledge points covered by them so that the user can get the appropriate questions convenient. And then achieve a better learning efficiency.
In this thesis, we continue studies which the classification system of K-12 learning materials. In addition to planning database for learning materials, and proposed cross-topic classification system. The traditional way of learning is often for each different single point of knowledge to learn. In the original system for such problems have a good classification performance. Some question of the large entrance exam and advanced question have the different concept of cross-topic. Therefore, we extend the original Convolutional Neural Network (CNN) and support vector machine (SVM) hybrid classifier and proposed multi-label classification model for cross-topic questions. Finally, we compare the strategies proposed by classification system studies of K-12 learning materials with our multi-label classification model. The experiment shows that the multi-label classification model can outperform original strategies of classification.

摘要 I
Abstract II
致謝 IV
Table of Contents V
List of Figures VII
List of Tables IX
1. Introduction 1
1.1 Background and Motivation 1
1.2 Challenge 9
1.3 Organization of Thesis 12
2. Related Works 13
2.1 CNN-SVM Hybrid Model 14
2.2 Label-based Classification 17
3. Multi-label CNN-SVM Hybrid Model 19
3.1 System Goal 19
3.2 System Overview 20
3.3 Preprocessing and Feature Extraction 22
3.3.1 Preprocessing 22
3.3.2 Feature Extraction 24
3.4 Multi-label CNN-SVM Hybrid Model 27
3.4.1 Training Phase 28
3.4.2 Predicted Testing Phase 30
4. Implementation 31
4.1 Learning Material Management System Overview 31
4.2 Database Design 33
5. Evaluation 35
5.1 Data Set 35
5.2 Evaluation Metrics 36
5.3 Multi-label CNN-SVM Hybrid Model Performance 37
5.4 Experimental Setup 39
5.4.1 Multi-label Strategy of Multi-Chapter Classification(MSMC) 39
5.4.2 Experiment 41
6. Conclusion and Future Works 42
List of References 44

[1] F. Kanwal and M. Rehman, "Factors Affecting E-Learning Adoption in Developing Countries-Empirical Evidence From Pakistan's Higher Education Sector," IEEE Access Volume: 5, 09 June 2017.
[2] D. Light and E. Pierson, "The Use of Khan Academy in Chilean Classrooms: Study of an Intel Funded Pilot Program in Chile," 2014 IEEE 14th International Conference on Advanced Learning Technologies, 2014.
[3] J. Jia and Q. Wang, "Using DACP (Data Analyzer of Coursera Platform) to Analyze MOOC Platform Data," 2015 International Conference of Educational Innovation through Technology, 2015.
[4] J. Liang, J. Yang and e. al, "Big Data Application in Education:Dropout Prediction in Edx MOOCs," IEEE Second International Conference on Multimedia Big Data, 2016.
[5] S. Penfold, "Top 10 eLearning Trends To Watch In 2017," eLearning INDUSTRY, 13 December 2016. [Online]. Available: https://elearningindustry.com/top-10-elearning-trends-to-watch-in-2017.
[6] Wattinger, D. P. Nguyen and etc., "Problem-Based Learning Using Mobile Devices," Advanced Learning Technologies, 2006. Sixth International Conference on, 05 July 2006.
[7] Ciresan, D. Claudiu and e. al, "Convolutional neural network committees for handwritten character classification," 2011 International Conference on Document Analysis and Recognition. IEEE, 2011.
[8] Y. LeCun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-Based Learning Applied to Document Recognition," Proceedings of the IEEE, Nov 1998.
[9] Lawrence, Steve and e. al, "Face recognition: A convolutional neural-network approach," IEEE transactions on neural networks 8.1, 1997.
[10] D. Meena and R. Sharan, "An Approach to Face Detection and Recognition," IEEE International Conference on Recent Advances and Innovations in Engineering, 23-25 December 2016.
[11] A. Buriro, Z. Akhtar, B. Crispo and F. D. Frari, "Age, Gender and Operating-hand Estimation on Smart Mobile Devices," 2016 International Conference o,f the Biometrics Special Interest Group (BIOSIG), 2016.
[12] P.-C. W. Huan-Huan Zhang, Y.-J. Wang, T. Ruan and H.-F. Wang, "Risk factors of heart failure for patients classification with extreme learning machine," 2016 International Conference on Machine Learning and Cybernetics (ICMLC), 2016.
[13] S. Gunn, "Support Vector Machines for Classification and Regression," technical report, Information: Signals, Images, Systems (ISIS)Research Group, Univ. of Southampton, 1998.
[14] Y. Kim, "Convolutional Neural Networks for Sentence Classification," Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing(EMNLP 2014), 2014.
[15] S.-Y. Ho, "A Self-relevant CNN-SVM Model for Problem Classification in K-12 Question-driven Learning," July 2016.
[16] M.-L. Zhang and Z.-H. Zhou, " A Review on Multi-Label Learning Algorithms," IEEE Transactions on Knowledge and Data Engineering, Aug 2014.
[17] T. Li and M. Ogihara, "Detecting Emotion in Music," In Proceedings of the International Symposium on Music Information Retrieval, pp. 239-240, 2003.
[18] S. Godbole and S. Sarawagi, "Discriminative Methods for Multi-labeled Classification," KReSIT, IIT Bombay Powai, Mumbai, 2004.