- 學位論文
以貝氏網路為基礎之能力指標測驗編製及補救教學動畫製作–以五年級數學領域幾何相關指標為例
Competence Indicators Test and Remedial Instruction Developments Based on Bayesian Networks – The Geometry Related Indicators of Mathematics in Grade 5
摘要
中文摘要 本研究擬以國小五年級數學「幾何」的能力指標為依據，嘗試採用以機率推理為基礎的貝氏網路作為分析工具，以探討應用貝氏網路於診斷學生錯誤類型的可行性。同時將子技能納入分析，探討哪些子技能的缺失造成這些錯誤類型的產生。發展一套以九年一貫能力指標為基礎的智慧型線上測驗及補救教學系統，讓學生透過網路參加測驗並了解自己的測驗結果。當學生測驗完畢後，即時呈現作答結果、學科概念的診斷情形，並根據錯誤類型的分布狀況，進行線上補救教學。 因此，本研究主要有四個目的 一、探討國小五年級數學領域「幾何」相關能力指標錯誤類型的發生情形。 二、應用貝氏網路於分析國小五年級幾何相關能力指標錯誤類型，並建置測驗試題及貝氏網路架構圖。 三、建立一套以錯誤類型為基礎之電腦補救教學動畫。 四、驗證電腦化適性補救教學動畫是否達到效果。 根據本研究施測流程進行以及施測結果分析，得到以下結論： ㄧ、貝氏網路模式用於診斷「幾何」能力指標錯誤類型的可行性，經實驗發現錯誤類型、子技能、能力指標有無分類的平均正確率均達九成以上，且動態分類決斷值的平均分類辨識率優於固定分類決斷值的平均分類辨識率。因此，使用動態選取分類決斷值法來提高辨識率做為貝氏網路推論學生診斷結果更可增加其可行性。 二、利用電腦多媒體對學生的吸引力，給予適性且正確的補救教學後，後測皆優於前測且t檢定***p < .001達顯著水準，可見本系統之電腦補救教學對於澄清受試者的幾何概念有立即的成效且成效顯著。 關鍵詞：貝氏網路、電腦化診斷測驗、電腦化補救教學。
並列摘要
Abstract The main idea of the study is to research the ability of geometry on Grade 5 and demonstrate the applicable of students mistaken types based on Bayesian Networks which is a probability analysis method. Students attend the test through the Learning Educational Program online, which is developed based on the index of Grade 1-9 Curriculum. The system can show the subject comprehension and begin to the Learning Educational Program on the basis of the mistaken type distribution. There are four main purposes of the study as below. 1. Demonstrating the geometry index of mistaken types on Grade 5. 2. Appling the Bayesian Networks to analyze geometry index of mistaken types on Grade 5, set up the quiz question and Bayesian Networks Framing. 3. Establishing the Learning Educational Program based on the mistaken types. 4. Demonstrating the effective of the computerized adaptive flash. According to the analysis of the test result by Bayesian Networks Model, the average corrective is up to 90% with mistaken types, sub-skills and capable index. The first conclusion shows it is more adaptive by using dynamic classified method to improve the recognizable degree on inferring the test result. After giving the suitable teaching to the students by the attracting flash, the students get higher scores than before.(t-test, p<0.001) The second conclusion of the study shows it is significantly effective to the students to clear the geometry concept by using the Learning Educational Program. Keywords: Bayesian belief networks, Computerized Diagnosing Test, Adaptive Remedy Instructions.