【摘要】：現(xiàn)階段,社會飛速發(fā)展,計算機技術(shù)不斷革新。在教育領(lǐng)域,互聯(lián)網(wǎng)+教育這一傳統(tǒng)行業(yè)成為一個新的熱點和藍海,相關(guān)交叉學(xué)科的專家根據(jù)這一課題展開了廣泛的研究。以共享知識資源為代表的在線課堂也發(fā)展的越來越快,譬如國內(nèi)的清華在線,超星泛雅平臺,以及國外的三大在線課程平臺Coursera、Udacity和edX等課程資源極其豐富,一些優(yōu)秀的教師也加入其中,帶來了優(yōu)質(zhì)的師資資源。但是由于平臺功能的限制,對于一些學(xué)生用戶的學(xué)習(xí)日志、學(xué)習(xí)路徑等行為數(shù)據(jù)沒有相關(guān)的數(shù)據(jù)分析,無法對個體建立偏好模型,教師很難了解每一名學(xué)生的學(xué)習(xí)能力以及學(xué)習(xí)風(fēng)格,因此不能針對性的制定個性化教學(xué)目標。由此,本文借助數(shù)據(jù)挖掘工具Weka,針對山東師范大學(xué)網(wǎng)絡(luò)學(xué)習(xí)平臺產(chǎn)生的大量教學(xué)數(shù)據(jù)、學(xué)生成績數(shù)據(jù)、學(xué)生學(xué)習(xí)日志等,展開具體的挖掘分析。具體研究目標如下:1、利用關(guān)聯(lián)算法,找到真正影響學(xué)生成績的因素,給教師提供教學(xué)質(zhì)量的分析和改進。2、利用相關(guān)的聚類分類算法分析學(xué)生學(xué)習(xí)能力,將具有相同或者相似風(fēng)格的學(xué)生組合到一起,統(tǒng)一進行教學(xué)目標管理,發(fā)現(xiàn)學(xué)生與學(xué)生之間的聯(lián)系。3、通過教學(xué)實踐以及與老師之間的交流反饋,提出了四種學(xué)習(xí)風(fēng)格數(shù)據(jù)量化指標,教師可以針對學(xué)生線上學(xué)習(xí)的數(shù)據(jù)區(qū)分不同學(xué)習(xí)風(fēng)格的學(xué)生,結(jié)合聚類分析的結(jié)果,進行相關(guān)的任務(wù)布置和管理,實現(xiàn)個性化教學(xué)。4、針對不同教師的數(shù)據(jù)挖掘需求,提出了一個成績分析平臺模塊框架,可以幫助教師針對學(xué)生的實際情況進行相關(guān)數(shù)據(jù)的挖掘,降低教師的教學(xué)成本。本文主要有兩個創(chuàng)新點:1、技術(shù)上的創(chuàng)新:改進以往教育領(lǐng)域通過調(diào)查問卷來進行教學(xué)分析的做法,充分利用算法的優(yōu)勢,進行數(shù)據(jù)量化,通過科學(xué)的算法來對數(shù)據(jù)進行處理和分析,使得整個數(shù)據(jù)分析嚴謹性、操作性比較強。2、模型上的創(chuàng)新:建立了數(shù)據(jù)挖掘模型,通過學(xué)生數(shù)據(jù)分析不同學(xué)習(xí)風(fēng)格的學(xué)生特點以及相關(guān)的量化指標,教師可以建立不同學(xué)習(xí)風(fēng)格的學(xué)生分組,從而制定個性化教學(xué)目標,提高學(xué)生的創(chuàng)新意識和合作意識。通過網(wǎng)絡(luò)教學(xué)平臺的數(shù)據(jù)分析,教師可以根據(jù)不同學(xué)生的風(fēng)格針對性的制定教學(xué)計劃,學(xué)生可以變被動接受新知識為主動學(xué)習(xí)新的學(xué)習(xí)資源,教師實現(xiàn)因材施教。
[Abstract]:At present, with the rapid development of society, computer technology is constantly innovating. In the field of education, the traditional industry of Internet education has become a new hot spot and blue sea. The online classroom, represented by shared knowledge resources, is also developing more and more rapidly. For example, Tsinghua online in China, Chaoxing Pan-ya platform, and three online courses platforms, Coursera,Udacity and edX, are extremely rich in resources. Some excellent teachers also joined in, bringing high-quality teacher resources. However, due to the limitation of platform function, there is no relevant data analysis for some student users' behavior data, such as learning log and learning path, so it is impossible to establish preference model for individuals. It is difficult for teachers to understand each student's learning ability and learning style. Based on the data mining tool Weka, this paper analyzes a large number of teaching data, student score data and student learning log generated by the network learning platform of Shandong normal University. The specific research objectives are as follows: 1. By using the correlation algorithm, we can find out the factors that really affect the students' achievement, and provide teachers with the analysis and improvement of the teaching quality. Second, we can use the related clustering classification algorithm to analyze the students' learning ability. Students with the same or similar styles together, unified teaching objectives management, found the relationship between students and students. 3, through teaching practice and exchange feedback with teachers, Four data quantification indexes of learning styles are put forward. Teachers can distinguish students with different learning styles according to the data of students' online learning. Combining the results of cluster analysis, teachers can arrange and manage relevant tasks. To realize individualized teaching. 4. According to different teachers' demand of data mining, this paper puts forward a module framework of achievement analysis platform, which can help teachers to mine relevant data according to students' actual situation and reduce teachers' teaching cost. There are two main innovations in this paper: one is technical innovation: to improve the past teaching analysis in the field of education through questionnaires, to make full use of the advantages of the algorithm, and to quantify the data. Through the scientific algorithm to process and analyze the data, make the whole data analysis rigorous, maneuverability relatively strong. 2. The innovation of the model: set up the data mining model, Through student data analysis of students' characteristics of different learning styles and related quantitative indicators, teachers can set up groups of students with different learning styles, so as to formulate individualized teaching objectives and improve students' consciousness of innovation and cooperation. Through the data analysis of the network teaching platform, teachers can make teaching plans according to different students' styles, students can change passive acceptance of new knowledge into active learning resources, and teachers can teach students according to their aptitude.
【學(xué)位授予單位】：山東師范大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP311.13

【參考文獻】

相關(guān)期刊論文 前10條

1 張明陽;常盼;徐冬冬;顧芝亞;王東林;陶陸陽;;獨立學(xué)院醫(yī)學(xué)生法醫(yī)學(xué)個性化教學(xué)模式的探究[J];教育教學(xué)論壇;2016年52期

2 包耕;張玲樂;;關(guān)聯(lián)規(guī)則隱藏算法綜述[J];軟件導(dǎo)刊;2016年11期

3 孫崴;劉學(xué)敏;許紅梅;;基于大學(xué)生認知風(fēng)格特征的MOOCs課程建設(shè)研究[J];現(xiàn)代教育科學(xué);2016年10期

4 杜唐;徐仕寶;李明東;;基于聚類分析算法的應(yīng)用性改進[J];信息通信;2016年10期

5 陳志飛;馮鈞;;一種基于Apriori算法的優(yōu)化挖掘算法[J];計算機與現(xiàn)代化;2016年09期

6 徐劉杰;;基于網(wǎng)絡(luò)學(xué)習(xí)平臺的翻轉(zhuǎn)課堂教學(xué)研究[J];軟件導(dǎo)刊(教育技術(shù));2016年07期

7 應(yīng)國良;;國際學(xué)校中“個性化教學(xué)”帶來的啟示——評《如何進行個性化教學(xué):來自國際學(xué)校的啟示》[J];中國教育學(xué)刊;2016年06期

8 陳俞;趙素云;陳紅;李翠平;孫輝;;統(tǒng)計粗糙集[J];軟件學(xué)報;2016年07期

9 李牧南;;基于關(guān)聯(lián)規(guī)則挖掘競爭情報研究前沿分析[J];情報雜志;2016年03期

10 王景中;張存正;;用于網(wǎng)絡(luò)行為分析的一種改進K-means算法[J];北方工業(yè)大學(xué)學(xué)報;2016年01期

相關(guān)碩士學(xué)位論文 前3條

1 鮑素貞;數(shù)據(jù)挖掘技術(shù)在個性化網(wǎng)絡(luò)教學(xué)平臺中的應(yīng)用研究[D];聊城大學(xué);2015年

2 滕子牧;數(shù)據(jù)挖掘關(guān)聯(lián)規(guī)則算法研究與應(yīng)用[D];遼寧科技大學(xué);2015年

3 高鵬;基于數(shù)據(jù)挖掘的個性化網(wǎng)絡(luò)教學(xué)平臺研究[D];西北大學(xué);2005年

，

本文編號：2201807