請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/59888
標題: | 智慧型英文出題系統之開發與效益分析 Development and Effectiveness Analysis of an Intelligent Online English Learning System |
作者: | Dai-Ling Shih 施岱伶 |
指導教授: | 孫雅麗(Yea-li Sun) |
關鍵字: | 自動出題,自然語言處理,線上測驗,自律學習,泛讀練習, automatic question generation,natural language processing,extensive reading,self-regulated learning,customized assessment,multiple-choice questions, |
出版年 : | 2016 |
學位: | 碩士 |
摘要: | 英文能力對於英文非母語(non-native)的學生來說是重要的核心競爭力之一,但是他們可能有一些學習英文上的困難,例如詞彙量的不足,或是文法不熟悉。因此,學者們提出許多方法來幫助他們學習。然而學者們也注意到除了協助程度較不好的學生學習之外,已經具有一定程度的學生也需要持續練習以維持他們的英文能力。在眾多的練習方法中,透過大量閱讀文章的泛讀(extensive learning)訓練不僅可以複習已習得的英文技能,也可以從閱讀不同文章中學習更多英文技巧。因此許多學者認為泛讀訓練對已經具有一定程度的學生而言是重要且最有用的方法之一。而泛讀能力的訓練方法中,又以在閱讀文章後給予測驗最常見,因為測驗不僅能展現學生的學習成果,也能讓他們在測驗中複習所學習的內容。然而練習過程中需要大量的文章,而針對大量文章出題往往造成教師的負荷。過去有許多研究基於自然語言處理(natural language processing)方法提出自動產生測驗的機制。相關研究顯示,這種方法不僅有效率也具有一定的正確性,另外也可以減輕教師出題的負擔。
有鑑於此,本研究嘗試提出一套具有自動出題機制的線上學習系統(AutoQuiz),協助教師及學生進行泛讀訓練。另一方面,許多學者認為在進行這類學習活動時,學習者的自律表現也會有一定程度的影響,因此本研究也將自律活動的引導機制建構於系統中。為了評估此系統的效益,我們規劃了二個實驗並藉此來探究此系統是否對學生學習有所影響、以及在這樣學習環境中自律學習的引導機制是否對於他們的學習表現帶來影響。實驗結果顯示,86%以上的學生認為使用AutoQuiz系統進行泛讀訓練對提升他們的英文能力有幫助,而加入自律學習引導機制後,學生的英文能力的進步程度與內在學習動機顯著多於沒有自律的學生。 English skills have been recognized as a core competence for non-native English speaking students. However, learning English is a difficult and challenging task for non-native English speaking students. Consequently, the development of effective learning strategies and environments for English courses has become an important issue. However, seldom of the existing learning strategies focus on the learners who have good proficiency, besides, learners who already had good English skills still need practices and appropriate learning strategies to maintain their proficiencies. Among all of the research dealing with this issue, extensive reading (ER) has gain a lot of interests in the past decade since it benefits learners not only on vocabulary acquisition but also spelling and increasing grammatical knowledge especially in higher education settings. Meanwhile, placing an assessment in the training of extensive reading is recognized as a better method for uncovering students’ learning status and support the training of extensive reading. However, it is unfeasible for teachers to prepare assessments for every article. As a result, many computer-assistant tools using natural language processing (NLP) technique are proposed to facilitate teachers in preparing the assessment for the training of ER. To this end, we implemented an online learning system with an automatic question generation mechanism and self-regulation functions to generate customizable assessments for supporting the training of ER. On the other hand, many researchers addressed that the self-regulated learning (SRL) is a factor which affects the learning performance of the learners. Therefore, the proposed system in this study also takes the self-regulated learning strategy into account. To identify the influences of the proposed strategy, an experiment was conducted in a real classroom to evaluate the effectiveness of the proposed approach by comparing the learning performance of the students who learned with the system which contains the self-regulated learning approach and those who learned with the system without the self-regulated learning approach. The experimental results show that the proposed approach not only improved the students’ learning performance, but also improved their inner motivation in learning English. Moreover, 86% of the participants agree that the proposed system supports them in the training of the extensive reading. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/59888 |
DOI: | 10.6342/NTU201700297 |
全文授權: | 有償授權 |
顯示於系所單位: | 資訊管理學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-105-1.pdf 目前未授權公開取用 | 3.68 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。