本研究探討國小高年級學生對於人工智慧應用之「推薦系統的發展」非形式推理思考表現，與其針對這個議題之線上合作論證表現，本研究的研究對象為某一國小五年級學生48名，研究工具採用Wu & Tsai （2007）非形式推理問卷，蒐集研究對象在人工智慧「推薦系統的發展」議題上的非形式推理的表現，並進行分析，再以人工智慧「推薦系統的發展」議題，進行20分鐘的線上合作論證，平台資料以McAliste等（2004）的合作論證對話架構，進行合作論證行為的質性分析。 主要的研究發現為： 一、 高低非形式推理能力的國小高年級學生在線上論證表現並無顯著差異。 二、 同質組在進行線上合作論證時，與全班和異質組更對組員所提出的論點有不同意的地方提出挑戰後進入認同，而異質組較同質組更能針對對方論點持續的進行挑戰，達到對雙方論點的了解。 三、 不同性別的分組相較於同性別的分組，更能在論點的「挑戰」後整合對方論點，提出可能成立的例證。 四、 「異質組」較「同質組」在線上合作論證上，不易出現非論證對話行為。 五、 「不同性別組」較「同性別組」在線上合作論證上，不易出現非論證對話行為。 本研究產出的結果，對於國小高年級進行合作論證的行為模式能有所了解，對於後續執行相關研究也能有所幫助。 ;This study explored the performance of the non-formal reasoning of the "development of recommendation system" in the application of artificial intelligence for the senior students of elementary schools. In contrast to the performance of online collaborative demonstrations on this topic, the object of this study is 48 fifth-grade students. 0ne 0f the research tool was based on Wu & Tsai (2007) Informal Reasoning Questionnaire, which collected and analyzed the performance of informal reasoning on the topic of artificial intelligence "recommendation system development", and then used artificial intelligence to "recommend system development issue". Then, a 20-minute online cooperation demonstration is conducted. The platform information uses the cooperation demonstration dialogue framework of McAliste et al. (2004) to conduct a qualitative analysis of the cooperation demonstration behavior. The main research results are as follows: 1. There is no significant difference in online demonstration performance between senior elementary school students with high and low informal reasoning abilities. 2. When the homogeneous group was conducted with online collaborative demonstrations, the whole class and the heterogeneous group challenge the arguments put forward by the group members and then enter into agreement, while the heterogeneous group is more able to continue to challenge the opponent′s arguments than the homogeneous group. To achieve an understanding of the arguments on both sides. 3. Compared with groups of the same sex, groups of different genders can integrate the arguments of the other party after the "challenges" of the arguments and provide some useful examples. 4. Compared with the "heterogeneous group" and the "homogeneous group" online collaborative argumentation, non-argumentation dialogue behavior is less likely to occur in the "heterogeneous group". 5. Compared with the "different gender group" and the "same sex group" online collaborative argumentation, non-argumentation dialogue behavior is less likely to occur in the "different gender group". The results of this study can provide an understanding of the behavioral patterns of cooperative demonstration in the upper grades of elementary schools, and can also be helpful for the subsequent implementation of related research.
