臺灣博碩士論文加值系統

近年來大學入學考試的招生入學方案屢屢實施重要的變革，主要的目的乃是為實現高中新課綱目標以符合高中課程教學的適性多元發展。在各項變革中，希望能將學科能力測驗考科數目減少以及重視參採學習歷程，藉此改變以往以考試為主要的學習模式，朝向適性多元學習模式。 本研究應用大學入學考試數據於大學個人申請中學生對於各校系組的申請與入學之分析。藉由考生最後所選擇的校系組作為依據，而類比各校系組的競賽而進行評比的方式。據此便可依考生最終選擇校系的結果而應用大學入學考試數據來建立整體的社群網路(social networks)及其相對的競賽量化評比。應用本研究所提評比於社會網路的分析不但可改善過去各大專院校在進行申請入學招生結果無法量化分析的困難與窘境，更可落實教育部所倡導的大學招生專業化，以及提供考生校系組選擇的參考。

Recently, the way of university entrance examinations has undergone significant changes. The aim is to implement the new curriculum to adapt to the development of teaching adaptability and diversity in high school. Among the reforms, the most significant things include the number of test subjects in the entrance examination is decreased, and the emphasis on the learning process of acquisition. The purpose is to modify the current exam mode of study as multiple adaptable learning modes. In this thesis, the university entrance exams data has been used to analyze the admission by applications of students to the specific departments of the universities. Based on the students’ final decisions to choose the universities, it looks like the competition between any pairs of universities to recruiting students. By using the university entrance exams data, the social networks topology is then constructed, and the related competition ranking can be then provided. It is expected that the proposed approach in combination with competition niversity entrance exams data and social network analysis can overcome the difficulties and dilemmas that the quantity analysis of university enrollment applications are hard to be done in the past. Moreover, our method can provide the help to Ministry of Education in promoting the implementation of university enrollment specialization to the universities, and provide reference for students to apply the departments of universities.

摘要................................i
Abstract............................ii
誌謝................................iii
目錄................................iv
表目錄..............................v
圖目錄..............................vi
第一章 緒論.........................1
1.1 研究背景........................1
1.2 研究問題........................1
1.3 研究目的........................2
第二章 文獻探討.....................3
2.1 多元入學管道....................3
2.2 個人申請入學校系競爭量化評比....4
2.2.1 排名系統......................4
2.2.2 雙人競賽系統..................5
2.2.3 TrueSkill系統.................6
2.3 社群網路........................7
第三章 研究內容與方法...............8
3.1 研究架構........................8
3.2 資料來源........................8
3.3 TrueSkill.......................9
3.4 社群網路........................10
第四章 研究結果.....................11
第五章 結論.........................28
5.1研究結論.........................28
5.2研究侷限與未來展望...............28
Extended Abstract...................31

[1]教育部 (2017), “大學多元入學升學網-考試方式,” http://nsdua.moe.edu.tw/index.php/examinations/examination-1
[2]彭杏珠(2017), “大學考招新制111學年上路 「X+P+Y」非懂不可,” 遠見雜誌特刊, 出版社：遠見天下, 2017/06/27.
[3]Rotou, O., Qian, X., & von Davier, M. (2015). Ranking systems used in gaming assessments and/or competitive games. ETS Research Memorandum Series, ETS RM-15-03.
[4]Schadenberg, B. R., Neerincx, M. A., Cnossen, F., & Looije, R. (2017). Personalising game difficulty to keep children motivated to play with a social robot: A Bayesian approach. Cognitive systems research, 43, 222-231.
[5] Tseng, W. T. (2016). Measuring English vocabulary size via computerized adaptive testing. Computers & Education, 97, 69-85.
[6] Pelánek, R. (2016). Applications of the Elo rating system in adaptive educational systems. Computers & Education, 98, 169-179.
[7] Glickman, M. E. (2012). Example of the Glicko-2 system. Boston University, 1-6.
[8] Glenz, K.H. (1980), Ingo system, In: Manfred van Fondern (ed.): Encyclopedia for Chess Friends, Publisher CJ Bucher, Lucerne and Frankfurt am Main 1980, ISBN 3-7658-0308-1, pp. 140-141.
[9] Elo, A. E. (1986), The rating of chess players, past and present, Ishi Press, New York.
[10] Edwards, R. (2017), “Edo historical chess ratings,” http://www.edochess.ca/index.html
[11] Graepel, T., & Herbrich, R. (2006). Ranking and matchmaking. Game Developer Magazine, 25, 34.
[12] Zermelo, E. (1929). The calculation of tournament results as a maximum-likelihood problem. Mathematische Zeitschrift, 29, 436-460.
[13] Bradley, R. A., & Terry, M. E. (1952). Rank analysis of incomplete block designs: I. The method of paired comparisons. Biometrika, 39(3/4), 324-345.
[14] Glickman, M. E. (2012). Example of the Glicko-2 system. Boston University, 1-6.
[15] Edwards, R. (2017), “Edo historical chess ratings,” http://www.edochess.ca/index.html
[16] Dangauthier, P., Herbrich, R., Minka, T., & Graepel, T. (2007). Trueskill through time: Revisiting the history of chess. Advances in neural information processing systems, 20.
[17] Kschischang, F. R., Frey, B. J., & Loeliger, H. A. (2001). Factor graphs and the sum-product algorithm. IEEE Transactions on information theory, 47(2), 498-519.
[18] Weng, R. C., & Lin, C. J. (2011). A Bayesian approximation method for online ranking. Journal of Machine Learning Research, 12(1).
[19] Graepel, T., & Herbrich, R. (2006). Ranking and matchmaking. Game Developer Magazine, 25, 34.
[20] Zhang, X., Graepel, T., & Herbrich, R. (2010, March). Bayesian online learning for multi-label and multi-variate performance measures. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (pp. 956-963). JMLR Workshop and Conference Proceedings.
[21] Graepel, T., Candela, J. Q., Borchert, T., and Herbrich, R. (2010), “Web-scale Bayesian clickthrough rate prediction for sponsored search advertising in Microsoft’s Bing search engine. In Proceedings of the 27th International Conference on Machine Learning 2010 (pp. 13–20).