結合智能倉儲排序桌遊之影像辨識的學習任務對大學生學習成效的影響

Abstract

本研究以智能倉儲排序為人工智慧應用主題，透過便利抽樣將範圍限定於臺灣北部某大學部大一學生，採用準實驗研究法，分別探討於人工智慧課程中是、否線上，及有、無結合智能倉儲桌上遊戲情境於學習影像辨識與排序之手機應用程式設計，對學生學習成效、學習動機和行為模式之影響。研究結果顯示三種方法的學習成效都有顯著提升，線上學習的學生有最佳的學習成效，推測可能因學生先備能力不錯可在線上個別學習，同時能按照自己觀看教材影片和內容的學習步調，同學除了網路繳交完成的手機程式，並有許多學生將實作過程進行螢幕錄影以確認這個程式是自己寫的，線上學習成效表現顯著優於線下學習。學習動機方面，線上學習的學生於內在目標導向、工作價值、學習的控制信念、學習和表現自我效能、價值成分、期望成分皆有提升。然而，推測可能使用桌遊實體授課需要更多的學習時間進行互動，故在控制相同時間限制下，與傳統學習成效沒有顯著差別，從傳統實體課程發現學生行為模式不習慣主動的探究學習，推論可能因學生對課程內容不熟悉導致於實體課程中傾向習慣於被指導。This study applied Smart-Warehouse sorting for learning the application of artificial intelligence. The participants were freshmen undergraduates in the northern Taiwan based on convenience sampling. This study adopted quasi-experimental research method. This study compared the learning effectiveness and motivations of the students learning online and those of the students learning off-line. Moreover, this study compared the learning effectiveness and motivations of the students using conventional off-line learning with those of the students applying Smart-Warehouse board game for smart phone application design so as to achieve image recognition and practice sorting.The results show that students who learned with three approaches made significant progress. The students learning online had the best learning outcome. It was inferred that students with enough prior proficiency learned individually and had their own learning pace for watching instructional video and material. In addition to hand over assignments of smart phone application online, many students record their implementation process so as to confirm that this program was written by themselves. The learning achievements of the students online was significantly higher than the learning achievements of the students offline. As for learning motivation, the intrinsic goal orientation, task value, control beliefs about learning, self-efficacy for learning and performance dimensions and value components and expectancy components were significantly promoted. However, it is inferred that the use of board game offline needs more learning time to carry out interaction. When the learning time is limited to be the same with the conventional offline learning, there is no significant differences between the learning achievements of board game offline learning and conventional offline learning. From the behavioral patterns of the conventional offline learning, it is found that the students were not used to active inquiry learning. It was inferred that students might tend to be taught in offline learning when they were not familiar with the learning content.