DC 欄位 |
值 |
語言 |
DC.contributor | 資訊工程學系 | zh_TW |
DC.creator | 李彣龍 | zh_TW |
DC.creator | Wen-Long Lee | en_US |
dc.date.accessioned | 2016-7-22T07:39:07Z | |
dc.date.available | 2016-7-22T07:39:07Z | |
dc.date.issued | 2016 | |
dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=103522020 | |
dc.contributor.department | 資訊工程學系 | zh_TW |
DC.description | 國立中央大學 | zh_TW |
DC.description | National Central University | en_US |
dc.description.abstract | 隨著雲端運算技術的成熟,數位學習利用其優點發展出各種應用。其中大規模開放式線上課程以及開放教育資源提供了豐富的教學資源,然而過多的教學資源將會造成資訊過載的問題,以及不適當教學資源結構所產生的學習迷失,學生無法決定該從何處學習、如何學習。因此可以利用知識地圖呈現知識架構來解決學習迷失,以及使用推薦系統幫助學生選擇適合的教學資源來解決資訊過載的問題。
本研究應用了目前熱門的大數據分析技術開發了適性化學習路徑推薦系統,使用Spark on Yarn的運算框架來建置分析環境。透過網路爬蟲收集教學資源、知識架構與專家詞庫,藉由教育專家詞庫中擷取教學資源的特徵,利用頻繁樣式探勘演算法建立關聯規則找出教學資源之關聯性。結合分類演算法分析教學資源的單元與難度,使得教學資源的關聯具有順序性。最後將分析結果使用知識地圖的方式視覺化的呈現在系統的Web端介面,並在知識地圖上規劃基本學習路徑提供學生參考。隨著學生使用紀錄、學習行為的資料逐漸增加,系統使用循序樣式演算法分析適合學生學習的路徑,達到適性化學習路徑的推薦。
本研究收集了「LearnMode學習吧」線上學習平台的國小數學教學影片,在沒有教育專家的幫助下,經過系統分析計算自動關聯在一起。透過資料視覺化產生知識地圖,使得學生容易發現教學影片之間的相關性。分析學生在系統上的使用歷程,並在知識地圖上推薦適合學生學習的路徑。本系統利用Chrome瀏覽器的plugin作為系統接口,強化「LearnMode學習吧」的功能,解決其學習迷失以及資訊負載的問題。 | zh_TW |
dc.description.abstract | In recent year, E-Learning uses Cloud Computing techniques to develop a variety of applications. Massive Open Online Courses and Open Education Resource provide a wealth of education resource. However, too many resources will lead to information overload. The inappropriate structure of E-Learning will cause learning disorientation and students are not easy to know where to learn and how to learn. Therefore, this research uses knowledge map and adaptive learning path to solving these problems. This research applies the popular technology of Big Data analytics to develop an adaptive learning path recommendation system. The system builds analysis environment with Spark on Yarn. Using the Web crawler to collect resources, knowledge structure, and expert vocabulary. This research combined with Frequent Pattern and Naive Bayes classifier to automatically establish the association of learning resource without the assistance of education expert. The analysis results are visualized and presented on the Web-based system. The Web-based system records the interaction log for each learner on the knowledge map. Through the sequential pattern mining to find out learner’s learning path. In addition, the system enhanced LearnMode by Chrome plugin. The research results show that applying Big Data analytics to develop an adaptive learning path recommendation system, and achieve adaptive learning and help learners to solve information overload and learning disorientation on LearnMode. | en_US |
DC.subject | 大數據分析 | zh_TW |
DC.subject | 適性化數位學習 | zh_TW |
DC.subject | 知識地圖 | zh_TW |
DC.subject | 學習路徑 | zh_TW |
DC.subject | 關聯規則 | zh_TW |
DC.subject | 分類 | zh_TW |
DC.subject | 推薦系統 | zh_TW |
DC.subject | Big Data analytics | en_US |
DC.subject | adaptive E-Learning | en_US |
DC.subject | knowledge map | en_US |
DC.subject | learning path | en_US |
DC.subject | association rule | en_US |
DC.subject | classification | en_US |
DC.subject | recommendation system | en_US |
DC.title | 應用大數據分析開發適性化學習路徑推薦系統 | zh_TW |
dc.language.iso | zh-TW | zh-TW |
DC.title | Applying Big Data analytics to develop an adaptive learning path recommendation system | en_US |
DC.type | 博碩士論文 | zh_TW |
DC.type | thesis | en_US |
DC.publisher | National Central University | en_US |