查詢結果分析
來源資料
相關文獻
- 生成式人工智慧下程式設計學習的新需求與新方法:以案例式推理為例
- 數位時代的造浪者:生成式人工智慧如何重塑圖書出版的內容行銷策略
- 教師在知識翻新環境之跨領域教學設計對GAI的提問與感知研究
- 學生程式設計能力影響因素之研究--以崑山技術學院資訊管理系為例
- Exposure: Giving Students Their Wings--A Revised Approach to Secondary School English
- C++Fuzzy類神經網路物件導向發展系統之建立
- 類比清晰度及系統性與程式設計概念之學習
- 中美視訊會議學習環境中互動性距離、滿意與學習結果之關係
- Designing Internet-based EFL Learning and Reconstructing Teacher-Student Interaction: A Case Study from the Course "Learning English through the Internet"
- WWW上智慧型合作學習環境之研究:一個全球氣候歸納探索學習環境
頁籤選單縮合
| 題 名 | 生成式人工智慧下程式設計學習的新需求與新方法:以案例式推理為例=New Demands and Methods for Programming Learning in the Era of Generative AI: A Case-Based Reasoning Approach |
|---|---|
| 作 者 | 莊永裕; 鄭丞傑; 蔡方慈; 王振漢; 陳國棟; | 書刊名 | 數位學習科技期刊 |
| 卷 期 | 17:3 2025.07[民114.07] |
| 頁 次 | 頁31-61 |
| 分類號 | 312.83 |
| 關鍵詞 | 人機協作; 生成式人工智慧; 案例式推理; 程式設計; 學習環境; Human-AI collaboration; Generative AI; Case-based reasoning; Programming; Learning environment; |
| 語 文 | 中文(Chinese) |
| DOI | 10.53106/2071260X2025071703002 |
| 中文摘要 | 隨著生成式人工智慧技術的進步,如何使用有效運用生成式人工智慧工具進行程式設計已成為重要議題。本研究針對程式初學者因缺乏基礎知識而導致對人工智慧工具過度依賴的問題,提出了一種基於案例式推理的學習模組。該模組旨在幫助學習者整合程式設計基礎概念與生成式人工智慧工具的使用技巧,從過往案例中汲取經驗,促進學習者的問題解決能力與自主學習能力。案例式推理學習模組結合了檢索、重用、修正和保留四步驟過程,引導學習者系統性地解決程式問題。本研究設計了13週的實驗,以臺灣某國立大學的非資訊相關學生為受試對象,透過前測、中測與後測,驗證案例式推理模組在提升學習者程式設計能力與生成式人工智慧工具使用效率方面的有效性。實驗結果顯示,學習模組增強學習者對程式設計概念的理解,也改善了學習者的學習行為與人工智慧工具的互動模式。相較於傳統學習方法,案例式推理模組能幫助學習者更好地理解程式設計的核心概念,並有效降低對人工智慧工具的依賴性。此外,受試者反饋指出,該模組可增強其解題信心,並促進其批判性思維與問題延伸能力。 |
| 英文摘要 | With advancements in Generative Artificial Intelligence (AI) technologies, effectively utilizing AI tools for programming has become a critical topic. This study addresses the issue of programming novices excessively relying on AI tools due to a lack of foundational knowledge. To tackle this, a learning module based on Case-Based Reasoning (CBR) was proposed. The module aims to help learners integrate fundamental programming concepts with AI tool usage skills, drawing insights from previous cases to enhance problem-solving abilities and self-directed learning. The CBR learning module incorporates four systematic steps: Retrieve, Reuse, Revise, and Retain, guiding learners to solve programming problems methodically. A 13-week experiment was conducted involving non-IT students from a national university in Taiwan. Pre-tests, mid-tests, and post-tests were employed to evaluate the effectiveness of the CBR module in improving learners' programming abilities and efficiency in using generative AI tools. The results demonstrated that the learning module enhanced learners' understanding of programming concepts and improved their learning behaviors and interactions with AI tools. Compared to traditional learning methods, the CBR module helped learners better grasp core programming concepts while effectively reducing their dependency on AI tools. Additionally, feedback from participants indicated that the module boosted their confidence in problem-solving, fostered critical thinking, and encouraged the exploration of extended problem scenarios. |
本系統中英文摘要資訊取自各篇刊載內容。