大學指定科目考試是進入大學的重要管道，在臺灣至少有40%的學生透過此管道進入大學，而建立一個大學入學考試的預測系統給學生在填寫志願時的建議是我們的目的。本篇論文中，介紹四種方法如何預測校系的錄取分數，而透過分析預測錄取分數以及真實錄取分數的差距，我們找到提升準確度的因素，並提出一個比其它方法更準確的預測方法。在實驗中，為了比較我們的方法與其它方法之預測能力，我們用三種分析值基於均方根誤差的計算方式，以及在資料檢索中使用的兩種分析值召回率和準確率。從2004到2014的實驗中，可以顯示出我們的方法比其它方法更加準確與有效。

Prediction systems for college entrance examination (CEE) are used to give some recommendations for students to choose their ideal department. In Taiwan, more than 40% students of senior high schools attend the CEE to enter a university/college. In this thesis, we first introduce four CEE prediction methods, and then propose an effective method based on the forward method. We analyze the differences between predicted scores and real scores to propose a factor to improve the performance. In addition, we compare these four prediction methods with ours, and explain how to apply our method to the prediction system. In the experiments, the CEE data from 2004 to 2014 are involved. We use three indicators based on the root mean square error (RMSE) and two indicators recall and precision to evaluate the performance of various methods. The experimental results show that our method outperforms the previous methods.

DISSERTATION VERIFICATION FORM i
ACKNOWLEDGEMENT iii
ABSTRACT iv
TABLE OF CONTENTS vi
LIST OF TABLES viii
Chapter 1. Introduction 1
Chapter 2. Preliminaries 3
2.1 The Forward Method 3
2.2 Forward Method with School Ranking 6
2.3 Lin’s Method 8
2.4 Prediction with Fuzzy Time Series 10
2.5 Comparison of Previous Methods 14
Chapter 3. An Improved Prediction Method for Entrance Scores 17
3.1 The Admission Rate 17
3.2 Our Prediction Method 18
3.3 Performance Evaluation 20
Chapter 4. Experimental Results 22
Chapter 5. Conclusion and Future Work 30
BIBLIOGRAPHY 31
Appendixes A. The Detail Results of Five Performance Indicators 33
Appendixes B. The Detail Results for Comparison of FMAR and Forward Method 35
Appendixes C. The Predicted MRTSs of Top 10 Departments in 2013 and 2014 38

[1] H. Al-Wazan, K. Ibraheem, and A.-G. Salim, “A hybrid algorithm to forecast enrolment based on genetic algorithms and fuzzy time series,” International Arab Journal of Information Technology, Vol. 11, No. 3, 2014.
[2] G. E. P. Box and M. E. Muller, “A note on the generation of random normal deviates,” The Annals of Mathematical Statistics, Vol. 29, No. 2, pp. 610–611, 06 1958.
[3] S.-M. Chen and C.-C. Hsu, “A new method to forecast enrollments using fuzzy time series,” International Journal of Applied Science and Engineering, Vol. 3, pp. 234–244, 2004.
[4] Y.-S. Chen, “The prediction of the department score of the college entrance examination in Taiwan,” Master’s thesis, National Sun Yat-Sen University, Kaohsiung, Taiwan, 2012.
[5] C.-L. Lin, “On the study of forecasting indices after the college entrance ex- amination with real data,” Master’s thesis, National Cheng Kung University, Tainan, Taiwan, 2008.
[6] M. Sah and K. Y. Degtiarev, “Forecasting enrollment model based on first-order fuzzy time series,” International Journal of Computer, Information Science and Engineering, Vol. 2, No. 1, 2007.
[7] Q. Song and B. S. Chissom, “Forecasting enrollments with fuzzy time series part I,” Fuzzy Sets and Systems, Vol. 54, No. 1, pp. 1–9, 1993.
[8] Q. Song and B. S. Chissom, “Fuzzy time series and its models,” Fuzzy Sets and Systems, Vol. 54, No. 3, pp. 269–277, 1993.
[9] Q. Song and B. S. Chissom, “Forecasting enrollments with fuzzy time series part II,” Fuzzy Sets and Systems, Vol. 62, No. 1, pp. 1–8, 1994.
[10] The College Entrance Examination Center, “CEEC.” http://www.ceec.edu.tw/, 1989.
[11] L. Zadeh, “Fuzzy sets,” Information and Control, Vol. 8, No. 3, pp. 338–353, 1965.
[12] M.-M. Zen, Z.-S. Chen, and H.-H. Jhan, “Statistical forecasting after the college entrance examination,” Journal of the Chinese Statistical Association, Vol. 46, No. 2, pp. 165–181, 2005.
[13] H. Zhang and X. Zhang, “Comments on data mining static code attributes to learn defect predictors,” IEEE Transactions on Software Engineering, Vol. 33, No. 9, p. 635, 2007.