依學生偏好及學習狀態建構之學習輔助者與知識協尋系統; A Learning Helper and Knowledge Finding System Based on Student Preference and Student Model

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/8433

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/8433

Title:	依學生偏好及學習狀態建構之學習輔助者與知識協尋系統;A Learning Helper and Knowledge Finding System Based on Student Preference and Student Model
Authors:	陳南光;Nan-kuang Chen
Contributors:	資訊工程研究所
Keywords:	資料探勘;學生偏好;學習輔助者;遠距教學;student preference;distance learning;data mining;expert finding
Date:	2000-06-26
Issue Date:	2009-09-22 11:26:51 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	透過網際網路進行遠距教學，已是未來不可避免之趨勢。在非同步遠距教學的環境中，學生難以從龐大的全班學生作品資料庫中找到其所需的資訊；而教師面對學生更多的課業問題或請求，本身的負荷增加；現有的討論區依關鍵字搜尋文章系統，使學生找到一大群文章後，仍需要檢視許多篇文章後，才能得到他所需的相關資訊。本論文的目標便是建立一個知識協尋系統，由學生作品集資料庫中找出符合學生需求之文章，並建立一個學習輔助者協尋系統，為學生找到能給予幫助的專家同學；以及為自動回答系統改進搜尋文章結果的排列順序，讓學生可以花更少的時間得到符合其需求的資訊。　　在主動推薦文章方面，我們根據學生對文章的偏好，從全班學生作品集資料庫中，採用k-Nearest Neighbor algorithm，找到和學生偏好相近的資訊，主動推薦給學生。　　在找尋專家同學方面，我們根據學生對其他同學的偏好，以及學生之學習者模式，採用Radial Basis Function Network，找到與學生偏好相近的專家同學。　　在改進搜尋文章排序方面，我們讓每篇文章的作者也成為在排序搜尋文章結果的依據：根據使用RBFN所找到專家同學的結果，給予每個專家同學一個Expert Weight，再加上原有向量空間模式的文章搜尋方式，來作為排序搜尋文章的依據。　　本篇論文建立了一個依學生偏好及學生學習狀態之學習輔助者及知識協尋系統，此系統輔助非同步遠距教學環境中之學生在全班學生作品資料庫中篩選出文章品質較佳之文章；在學生提出課業疑難時，提供在此課程概念上能夠給予幫助之專家同學；並改良搜尋相關資訊結果之排序，使學生減少檢視文章次數和時間，使其獲得符合其需求之文章。 Distance learning via Internet becomes more and more popular in recent years. In the asynchronous distance learning environment, students are hard to locate useful information from the large student knowledge database. Teachers’ load increases because of the increasing question from students. The result of searching documents by keyword make students need to examine many documents to get needed ones. Our goals are to establish an knowledge filtering system, find out the documents match a student’s need from the student knowledge database; and to establish a helper finding system, find a expert classmate for the student needs help; and to change the ranking of the searching result to make students reduce the number of times of examine documents. Based on document feedback from the student, we recommend the student the information from the student knowledge database using k-Nearest Neighbor algorithm. Based on classmate feedback from the student, we recommend the student the expert classmates using Radial Basis Function Network. We take the author of a document into consideration when ranking the searching result. We establish a learning helper and knowledge finding system based on students’ feedback and student model, it assists students in the asynchronous distance learning environment to find the high quality documents from the student knowledge database; when a student issue a question, it finds a suitable expert classmate to help this student; and it changes the ranking of the searching result to make students reduce the number of times of examine documents.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Size	Format

社群 sharing

Loading...