以人工智慧聚類演算法分析學科學習成績和六子棋之棋力關聯分析

The Correlation Study of the Grades to the Chess Ability of Connec6 through Optimized Clustering Algorithm

陳珮琪(Pei-Chi Chen)；萬絢(Shian Wan)；陳光(Kuang Chen)

嶺東學報

45期（2019 / 12 / 01）

155 - 172

繁體中文

本研究在探討棋藝能力與學科能力之相關性。六子棋人工智慧遊戲，是國立交通大學資訊工程系吳毅成教授發展出的策略性人工智慧遊戲，具備「規則簡單」、「變化複雜」、「遊戲公平」之特性，由於規則簡單、易於入手，故受測者無須先備知識，僅需了解遊戲規則即可立即受測，然而受測者若想贏棋，落子對奕時，卻需具備一定程度的觀察、思考、推斷能力，同時，此遊戲為人工智慧遊戲，可客觀地進行棋藝訓練與能力評估，十分適合於研究時使用。本研究利用六子棋人工智慧遊戲與學生們對弈，區分學生的六子棋棋力；並以六次學科成就評量，檢視五大學科之平均成績，因檢測學生學生學科成績複雜且不確定性相當高再用「粗集合分類法」篩選出重要屬性，並以「聚類優化」分析相關性，檢視六子棋棋力與五大學科學習成績之相關性，以做為棋藝能力暨學科能力相關性之評估參考。

The goal of this study is to find the relations between chess ability and subject performance. The six-game chess artificial intelligence game is a strategic artificial intelligence game developed by Professor Wu Yicheng of the Department of Information Engineering of National Chiao Tung University. It has three characteristics of ＂simple rules＂, ＂complex changes＂ and ＂game fairness＂. Because the rules are simple and easy to start, the tester do not need to have prior knowledge, they only need to know the rules of the game to be tested immediately. However, if the tester wants to win the game, the player needs to have a certain degree of observation, thinking and inference ability. For artificial intelligence games, the training and ability assessment can be performed objectively which is very suitable for research use. This study uses the six-game wise intelligence game to compete with the students to distinguish the six-song chess ability of the students. It is decided to evaluate the average scores of the five university subjects with six academic achievement evaluations. Then, we use the ＂rough set classification＂ to extract out the important attributes. We also analyzed the correlation with ＂cluster optimization＂ to examine the correlation between the six chess players and the scores of the five major courses as a reference for the evaluation of the chess ability and subject ability.

1. 教育部全球資訊網「各領域課程綱要」（http://www.edu.tw/）。
2. 國立交通大學實驗室「六子棋介紹」（http://www.connect6.org/）
3. Maruyama, T,Kawachi, T,Singh, VP(2005).Entropy-based assessment and clustering of potential water resources availability.Journal of Hydrology,309,104-113.
4. Salvi G.(2006).Segment boundary detectionvia class entropy measurements in connectionistb phoneme recognition.Speech Communication,48,1666-1676.
5. Vorko, A,Jovic, F.(2000).Multiple attribute entropy classification of school-age injuries.Accident Analysis & Prevention,32,445-454.
6. 柳宗林(2009)。嶺東科技大學。
7. 胡嘉芸(2012)。新竹市，國立交通大學。
8. 陳志宏(2011)。台北市，國立臺灣師範大學。
9. 陳勇年(2012)。嶺東科技大學。
10. 陳靖平(2009)。新竹市，國立交通大學。
11. 黃沛群(2008)。台中市，國立中興大學。
12. 詹宜智(2009)。新竹市，國立交通大學。
13. 劉思源(2006)。花蓮縣，國立東華大學。
14. 劉雲青(2009)。台北市，國立臺灣師範大學。
15. 蔡心迪(2011)。新竹市，國立交通大學。
16. 鄭吉閎(2012)。新竹市，國立交通大學。