臺灣博碩士論文加值系統

近年來由於人工智慧技術的快速發展，機器學習理論成為熱門研究的議題，並已廣泛應用於資料探勘、特徵識別及模糊邏輯系統等。本論文將適性評量機制分別應用於電腦對局領域、適性評量雛型系統驗證及高雄市國民中小學學生學習診斷與進展評量系統(Program of Learning Diagnosis and Progress Assessment for Primary and Secondary Students of Kaohsiung technology-based testing, POLDPA-tbt)之最大負載測試。在電腦對局領域應用方面，本論文基於模糊C均值(Fuzzy C-Means, FCM)分群演算法及模糊標記語言(Fuzzy Markup Language, FML)，將其應用於與電腦圍棋程式對奕之棋士棋力評估，進而將評估結果回饋給棋士，以增加棋士與電腦圍棋程式對弈之樂趣。在適性評量雛型系統驗證方面，本論文以試題反應理論(Item Response Theory, IRT)為基礎，根據受試者目前作答反應挑出符合受試者目前估計能力值之試題成為下一道試題。於系統負載測試方面，本論文首先使用模糊標記語言建置推論高雄市國民中小學學生學習診斷與進展評量系統之負載人數相關知識庫及規則庫。然後，本論文經由嚴謹的測試計畫及測試腳本，進行高雄市國民中小學學生學習診斷與進展評量系統之負載測試及系統功能驗證與確認(Verification and Validation, V&V)。接著，我們根據收集到的資料及建置之知識庫與規則庫，推論能夠成功完成適性評量測試之負載人數。最後，我們採用基因演算法(Genetic Algorithm, GA)以最佳化推論結果。實驗結果顯示，學習後的知識庫及規則庫其表現優於學習前的推論結果。未來，本論文將導入第二型模糊集合概念以更正確推論系統最大負載人數。

Due to the rapid development of artificial intelligence techniques, machine learning theory has become a popular research topic, and widely applied to data mining, feature recognition, fuzzy logic system, and so on. This thesis applies an adaptive assessment mechanism to the computer Go program, the developed adaptive assessment prototype’s verification, and the Program of Learning Diagnosis and Progress Assessment for Primary and Secondary Students of Kaohsiung technology-based testing (POLDPA-tbt) system load. For computer Go application, we use fuzzy C-Means clustering algorithm and fuzzy markup language (FML) to assess the rank of the invited Go players. Meanwhile, we also feedback the estimated rank to the Go players to increase the fun of playing with computer Go program. For the adaptive assessment prototype’s verification, this thesis uses an item response theory (IRT) to select an item whose difficulty fits with this examinee’s ability for his/her next time according to his/her current response. For the POLDPA-tbt system load, this thesis first uses FML to establish the knowledge base and rule base of POLDPA-tbt’s system load. Then, we use some strict test plans and test scripts to do many load tests and verification & validation (V&V) for this system. Next, we infer the number of students who are able to successfully finish the adaptive testing according to the collected data and the established knowledge base and rule base. Finally, we adopt a genetic algorithm to optimize the defuzzied results. The experimental results show that the after-learning knowledge base and rule base outperform the before-learning ones. In the future, we will try to introduce type-2 fuzzy set (T2 FS) to much accurately infer the system’s load.

目錄 V
圖目錄 VII
第一章 緒論 10
1.1 研究背景 10
1.2 研究動機 10
1.3 章節簡介 11
第二章 相關研究與文獻探討 13
2.1 模糊分群演算法 13
2.2 試題反應理論 15
2.3 模糊理論 18
2.4 模糊標記語言 21
2.5 軟體測試 24
2.6 基因演算法 25
第三章 圍棋適性評估系統驗證機制 28
3.1 適性版台灣魔圍棋簡介 28
3.2 圍棋適性評估系統架構 28
3.3 模擬次數分群建構機制 29
3.4 模糊集合建構機制 30
3.4.1 權重模糊變數 30
3.4.2 模糊推論知識庫 32
第四章 電腦適性評量雛型系統建置 35
4.1 電腦適性測驗簡介 35
4.2 電腦適性測驗系統架構 35
4.3 三參數對數模型試題參數建置機制 38
4.4 試題選題機制 44
第五章 電腦適性評量系統效能驗證機制 47
5.1 高雄市國民中小學學生學習診斷與進展評量系統簡介 47
5.2 評量系統網路連線路由架構及評量流程 48
5.3 系統負載驗證與確認 51
5.3.1 國家高速網路中心4月19號壓力測試案例 58
5.3.2 國家高速網路中心4月29號壓力測試案例 60
5.3.3 國家高速網路中心4月30號壓力測試案例 60
5.3.4 高雄市國民中小學協助實施壓力測試案例 61
5.4 模糊推論機制 61
第六章 實驗結果 68
6.1 圍棋適性評估系統驗證結果 68
6.2 電腦適性評量雛型系統 72
6.3 電腦適性評量系統效能驗證結果 76
6.3.1 國家高速網路中心4月19號壓力測試結果分析 76
6.3.2 國家高速網路中心4月29號壓力測試結果分析 78
6.3.3 國家高速網路中心4月30號壓力測試結果分析 79
6.3.4 高雄市國民中小學協助實施壓力測試結果分析 79
6.4 基因學習實驗結果分析 80
6.5 基因學習前與學習後實驗比較 85
6.4.1 基因學習前後MSE比較 85
6.4.2 適應值變動分析 87
第七章 結論及未來研究方向 91
7.1 結論 91
7.2 未來研究方向 92
Reference 94
附錄A 評量系統壓力測試計畫書 98
附錄B 評量系統壓力測試實際測試計畫書 105
附錄C 國家高速網路與計算中心4月19號壓力測試結果 111
附錄D 國家高速網路與計算中心4月29號壓力測試結果 113
附錄E 國家高速網路與計算中心4月30號壓力測試結果 114
附錄F 評量系統壓力測試實測結果 115
附錄G 評量系統壓力測試彙整資料 116
附錄H 評量系統壓力測試之模糊推論規則庫 119
附錄I 高雄市評量系統壓力測試每分鐘登入人數 126
附錄J 基因學習結果(Generation 3000) 129

[1]李允中，王小璠，蘇木春，”模糊理論及其應用”, 全華科技圖書股份有限公司，2012。
[2]李允中，”軟體工程”，社團法人台灣軟體工程學會，2013。
[3]吳孟珍，第二型基因模糊推論系統於圍棋棋力適性評估之應用，國立臺南大學資訊工程系研究所碩士論文，2013。
[4]余民寧，”試題反應理論(IRT)及其應用”，心理出版社，2009。
[5]吳裕益，陳英豪，”測驗與評量”，復文書局，1990。
[6]林豐澤，”演化式計算下篇：基因演算法以及三種應用實例”，智慧科技與應用統計學報，第三卷，第一期，pp. 29-56, 2005年6月。
[7]S. Alam, G. Dobbie, and P. Riddle, “An evolutionary particle swarm optimization algorithm for data clustering,” IEEE Swarm Intelligence Symposium, pp. 1-6, Sept. 2008.
[8]J. Chen and H. Zhang, “Research on application of clustering algorithm based on PSO for the web usage pattern,” International Conference on Wireless Communications, Networking and Mobile Computing, pp. 3705-3708, Sept. 2007.
[9]M. H. Wang, C. S. Wang, C. S. Lee, S. W. Lin, and P. H. Hung, “Type-2 fuzzy set construction and application for adaptive student assessment system,” 2014 World Congress Computational Intelligence (IEEE WCCI 2014), Beijing, China, Jul. 6-11, 2014.
[10]J. Bezdek, Pattern Recognition with Fuzzy Objective Function Algorithms, New York: Plenum, 1981.
[11]T. C. Havens, J. C. Bezdek, C. Leckie, L. O. Hall, and M. Palaniswami, “Fuzzy c-means algorithms for very large data,” IEEE Transactions on Fuzzy Systems, vol. 20, no. 6, pp. 1130-1146, Dec. 2012.
[12]S. Bandyopadhyay, A. Mukhopadhyay, and U. Maulik, “An improved algorithm for clustering gene expression,” dataBioinformatics, vol. 23, no. 21, pp 2859-2865, Aug. 2007.
[13]R. D. Bock, “A brief history of item theory response,” Educational Measurement: Issues and Practice, vol. 16, no. 4, pp. 21-33, 1997.
[14]R. K. Hambleton and R. W. Jones, “An NCME instructional module on: Comparison of classical test theory and item response theory and their applications to test development,” Educational Measurement: Issues and Practice, vol. 12, no. 3, pp. 38-47, 1993.
[15]C. S. Lee, M. J. Wu, M. H. Wang, O. Teytaud, H. M. Wang, and S. J. Yen, “T2FML-based adaptive assessment system for computer game of Go,” 2013 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2013), Hyderabad, India, Jul. 7-10, 2013.
[16]F. Doctor, H. Hagras, and V. Callaghan, “A fuzzy embedded agent-based approach for realizing ambient intelligence in intelligent inhabited environments,” IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, vol. 35, no. 1, pp. 55-65, Jan. 2005.
[17]G. Acampora and V. Loia, “Fuzzy control interoperability and scalability for adaptive domotic freamwork,” Data & Knowledge Engineering, vol. 1, no. 2, pp. 97-111, May 2005.
[18]G. Acampora, V. Loia, C. S. Lee, and M. H. Wang, On the Power of Fuzzy Markup Language. Berlin, Germany: Springer-Verlag, Jan. 2013.
[19]J. M. Mendel, Uncertain Rule-Based Fuzzy Logic Systems: Introduction and New Directions, Prentice Hill, 2001.
[20]J. Yen and R. Langari, Fuzzy Logic Intelligence, Control, and Information, Prentice-Hall, USA, 1999.
[21]C. S. Lee, M. H. Wang, C. H. Huang, S. L. Cho, C. S. Wang, B. H. Tsai, J. L. Yu, C. H. Chang, P. J. Hsieh, T. M. Chen, and C. H. Li, “FML-based website performance verification mechanism for adaptive assessment system,” The 2nd International Symposium on Advanced Intelligent Maritime Safety and Technology, Mokpo, Korea, May 15-17, 2014.
[22]C. H. Wang, T. P. Hong, and S. S. Tseng, “Integrating fuzzy knowledge by genetic algorithms,” IEEE Transactions on Evolutionay Computation, vol. 2, no. 4, pp. 138-149, Nov. 1998.
[23]C. H. Wang, M. B. Chang, T. P. Hong, and S. S. Tseng, “Integrating multiple rule sets by genetic algorithms,” IEEE International Conference on Systems, Man, and Cybernetics, Oct. 11-14,1998, vol. 2, pp.1524-1529
[24]C. S. Lee, S. M. Guo, and C. Y. Hsu, “A novel fuzzy filter for impulse noise removal,” Lecture Notes in Computer Science, vol. 3174/2004, Springer-Verlag, 2004.
[25]C. S. Lee, M. H. Wang, H. Hagas, Z. W. Chen, S. T. Lan, S. E. Kuo, H. C. Kuo, and H. H. Cheng, “A novel genetic fuzzy markup language and its application to healthy diet assessment,” (SCI) International Journal of Uncertainty, Fuzziness, and Knowledge-Based Systems, vol. 20, no. 2, pp. 247-278, Oct. 2012.
[26]C. C. Hung, S. Kulkarni, and B. C. Kuo, “A new weighted fuzzy c-means clustering algorithm for remotely sensed image classification,” IEEE Journal of , Selected Topics in Signal Processing, vol. 5, no. 3, pp. 543-553, Jun. 2011.
[27]M. Bristow, K. Erkorkmaz, J. P. Huissoon, S. Jeon, W. S. Owen, S. L. Waslander, and G. D. Stubley, “A control systems concept inventory test design and assessment,” IEEE Transactions on Education, vol. 55, no. 2, pp. 203-212, May. 2012.
[28]T. Ishii, P. Songmuang, and M. Ueno, “Maximum clique algorithm and its approximation for uniform test form assembly,” IEEE Transactions on Learning Technologies, vol. 7, no. 1, pp. 83-95, Jan. 2014.
[29]H. D. Huang, C. S. Lee, M. H. Wang, and H. Y. Kao, “IT2FS-based ontology with soft-computing mechanism for malware behavior analysis,” (SCI) Soft Computing, vol. 18, no. 2, pp. 267-284, Feb. 2014.
[30]C. S. Lee, M. H. Wang, M. J. Wu, O. Teytaud, and S. J. Yen, “T2FS-based adaptive linguistic assessment system for semantic analysis and human performance evaluation on game of Go,” (SCI) IEEE Transactions on Fuzzy Systems, 2014. (Accepted) (DOI: 10.1109/TFUZZ.2014.2312989)
[31]C. S. Lee, M. H. Wang, M. J. Wu, Y. Nakagawa, H. Tsuji, Y. Yamazaki, and K. Hirota, “Soft-computing-based emotional expression mechanism for game of computer Go,” (SCI) Soft Computing, vol. 17, no. 7, pp. 1263-1282, Jul. 2013.
[32]C. H. Liu, C. S. Lee, M. H. Wang, Y. Y. Tseng, Y. L. Kuo, and Y. C. Lin, “Apply fuzzy ontology and FML to knowledge extraction for university governance and management,” Journal of Ambient Intelligence and Humanized Computing, vol. 4, no. 4, pp. 493-513, Aug. 2013.
[33]A. Arcuri, “A theoretical and empirical analysis of the role of test sequence length in software testing for structural coverage,” IEEE Transactions on Software Engineering, vol. 38, no. 3, pp. 497-519, May. 2012.
[34]Y. Yin, B. Liu, and H. Ni, “Real-time embedded software testing method based on extended finite state machine,” Journal of Systems Engineering and Electronics, vol. 23, no. 2, pp. 276-285, Apr. 2012.
[35]Y. Yin, B. Liu, D. Zhong, and T. Jiang, “On modeling approach for embedded real-time software simulation testing,” Journal of Systems Engineering and Electronics, vol. 20, no. 2, pp. 420-426, Apr. 2009.
[36]C. F. Almeida, and N. Kagan, “Using genetic algorithms and fuzzy programming to monitor voltage sags and swells,” IEEE Intelligent Systems, vol. 26, no.2, pp. 46-53, Mar. 2011.
[37]L. X. Medeiros, G. A. Carrijo, E. L. Flores, and A. C. P. Veiga, “Genetic algorithms applied in face recognition,” IEEE (Revista IEEE America Latina) Latin America Transactions, vol. 10, no. 6, pp. 2280-2285, Dec. 2012.
[38]I. Ciornei, and E. Kyriakides, “Hybrid ant colony-genetic algorithm (GAAPI) for global continuous optimization,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 42, no. 1, pp. 234-245, Feb. 2012.
[39]K-12 Education Administration, Ministry of Education-Project for the implementation of remedial instruction, 2014 [Online] Available: http://priori.moe.gov.tw/index.php?mod=resource/index/content/material_math.
[40]Reading Chinese Classics Rules for Children, 2014 [Online] Available: http://edu.ocac.gov.tw/interact/ebook/classics/.
[41]均一教育平台，2014 [Online] Available: http://www.junyiacademy.org/.
[42]R. K. Hambleton, and H. Swaminathan, Item Response Theory: Principles and Applications, Springer Science & Business Media, 1985.
[43]D. C. Haley, Estimation of the dosage morality relationship when the dose is subject to error, Technical Report No. 15, 1952.