基於問答網路模型之中小學數位學習伴侶聊天機器人;K12 E-Learning Companion Chatbot Based on QANet Model

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/77645

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/77645

Title:	基於問答網路模型之中小學數位學習伴侶聊天機器人;K12 E-Learning Companion Chatbot Based on QANet Model
Authors:	王威凱;Wang, Wei-Kai
Contributors:	資訊工程學系
Keywords:	數位學習;聊天機器人;孤立感;冷淡感受;問答網路模型;E-Learning;Chatbot;Isolation;Impersonal;QANet
Date:	2018-07-26
Issue Date:	2018-08-31 14:51:35 (UTC+8)
Publisher:	國立中央大學
Abstract:	數位學習在近幾年來逐漸普及且愈趨流行。對於年輕世代的學生來說，使用行動平台或行動裝置已經是非常習以為常的習慣了。而基於行動平台的數位學習具有非常多優勢，使用者可以隨時隨地進行學習並自由管理學習進度及依據自身學習習慣做學習上的調整。然而近期的研究顯示在多數數位學習平台上，因先天缺乏與人互動的機制，而可能導致使用者產生孤立感以及冷淡感受。這樣的孤立感以及冷淡感受可能會降低數位學習平台使用者的學習積極性。本篇論文提出了一套基於問答網路模型的聊天機器人，用於輔助數位學習使用者。我們的聊天機器人主要設計作為具有與人類相似回應風格及回應能力的數位學習夥伴，除了作為單純的夥伴外，我們的聊天機器人也具有回應簡單數位學習課程相關的問題。實驗以及問卷結果顯示相較於我們聊天機器人設計所參考的數位學習平台，其提供的電話以及線上交談老師輔導服務，我們的聊天機器人更能大幅地減輕使用者在使用數位學習平台時所感受到的孤立感以及冷淡感受。本篇Paper是基於我們所投稿之期刊論文 [1]並專注在數位學習問題與問答網路模型(QANet)上。 ;E-Learning has become more and more popular in recent years. For young generation students, using mobile platform or devices is a familiar thing to do. E-Learning based on mobile platform has a lot of benefits. User can learn at any time, at anywhere, freely manage their learning progression and adapt with their own learning styles. However, recent studies show that E-Learning could causes isolation and impersonal feelings to user due to the nature of lacking human-like interactions in most of E-Learning platform. These feelings could lead to reduction of user’s motivation in learning. In this paper, a chatbot based on QANet model is proposed to reduce isolation and impersonal feelings from user alongside E-Learning platform. Our chatbot is mainly designed as a E-Learning companion which acts and responses like a real human companion. Other than that, our chatbot could reply some simple E-Learning course related questions. Experiment and Questionnaire results show that our chatbot can reduce users’ feeling of isolation and impersonal which performs better than the online texting-based and telephone-based teacher counselling services in the E-Learning platform our chatbot based on. This paper is based on our summited journal paper [1] and focused on E-Learning issues and QANet Model.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	145	View/Open

社群 sharing

Loading...