楊一柳
摘要:目前許多智能課件中的學(xué)生模型存在的問(wèn)題是缺乏對(duì)學(xué)習(xí)者學(xué)習(xí)情緒的獲取,教學(xué)策略的選擇缺乏人性化。針對(duì)此不足,依據(jù)貝葉斯概率推理建立以學(xué)習(xí)者學(xué)習(xí)情緒為中心的學(xué)生模型,它是一個(gè)動(dòng)態(tài)模型,該模型的特點(diǎn)在于能夠及時(shí)獲得學(xué)習(xí)者對(duì)智能課件的操作界面中某元素的感興趣程度,從而分析并動(dòng)態(tài)生成有利于學(xué)習(xí)者情緒優(yōu)化的教學(xué)策略,并作實(shí)驗(yàn)加以證明。
關(guān)鍵詞:貝葉斯;情緒獲??;學(xué)生模型;智能課件;教學(xué)策略
中圖分類(lèi)號(hào):TP391文獻(xiàn)標(biāo)識(shí)碼:A文章編號(hào):1009-3044(2012)01-0103-03
Analysis of Bayesian Reasonings Application of Students Model in Intelligent Courseware
YANG Yi-liu
(Teaching and Research Institute of College Computer Bohai University, Jinzhou 121013,China)
Abstract:The problems occurred in student models in many current intelligent courseware constitute lack of learning emotional acquisi? tion and short of humanity in the process of choosing teaching strategy. For this reason, learning emotional centered student models are es? tablished based on Bayesian probability reasoning. This dynamic model is featured by quick acquisition of learnersdifferent degrees of in? terest in elements of intelligent courseware interface and prompt analysis. Therefore learner emotional optimization teaching strategy will be generated dynamically. Experiment is undertaken as prove.
Key words:bayesian; emotion acquisition; students model; intelligent courseware;teaching strategy
人們根據(jù)不確定性信息做出推理和決策需要對(duì)各種結(jié)論的概率做出估計(jì),這類(lèi)推理稱(chēng)為概率推理。概率推理既是概率學(xué)和邏輯學(xué)的研究對(duì)象,也是心理學(xué)的研究對(duì)象。其中貝葉斯概率推理的問(wèn)題是條件概率推理問(wèn)題,在概率論的基礎(chǔ)上進(jìn)行不確定推理,是基于概率的一種算法,由一位偉大的數(shù)學(xué)大師Thomas·Bayes所創(chuàng)建的,這一領(lǐng)域的探討對(duì)揭示人們?cè)诟怕市畔⒌恼J(rèn)知加工過(guò)程與規(guī)律、指導(dǎo)人們進(jìn)行有效的學(xué)習(xí)和判斷決策都具有十分重要的理論意義和實(shí)踐意義。
智能課件[1]中存在著各種各樣的不確定性,針對(duì)學(xué)習(xí)者的多樣性、知識(shí)網(wǎng)絡(luò)結(jié)構(gòu)及信息的多樣性決定了認(rèn)知狀態(tài)診斷方法的不確定性,根據(jù)情緒數(shù)據(jù)進(jìn)行的關(guān)于教學(xué)策略的推理也是不確定的,而處理不確定性正是貝葉斯概率推理的優(yōu)勢(shì)。所以采用貝葉斯概率推理[2]作為學(xué)習(xí)者當(dāng)前情緒狀態(tài)與教學(xué)策略選擇的適應(yīng)度的診斷推理,也可以理解為教學(xué)策略對(duì)成功預(yù)見(jiàn)性的概率推理。
3結(jié)論
學(xué)習(xí)者當(dāng)前的情緒數(shù)據(jù)可以演繹出對(duì)當(dāng)前智能課件系統(tǒng)界面元素的適應(yīng)程度,是在不確定性前提下認(rèn)知診斷的方法。本文討論的在學(xué)生模型中利用貝葉斯概率推理對(duì)先驗(yàn)信息和后驗(yàn)信息的結(jié)合能力,對(duì)學(xué)生情緒的因果聯(lián)系進(jìn)行編碼,通過(guò)信息的不斷加入,對(duì)學(xué)生當(dāng)前學(xué)習(xí)情緒的評(píng)測(cè)和把握及時(shí)更新,為模型提供自適應(yīng)構(gòu)建方式和準(zhǔn)確實(shí)現(xiàn)智能課件所實(shí)施的教學(xué)策略(包括教學(xué)理論、界面元素設(shè)計(jì)、組織結(jié)構(gòu)等)提供良好的理論依據(jù)。
參考文獻(xiàn):
[1]陳曉丹,王建華.智能計(jì)算機(jī)輔助教學(xué)系統(tǒng)結(jié)構(gòu)模型的研究[J].哈爾濱師范大學(xué)自然學(xué)報(bào),2006,22(2): 68-70.
[2] Martin J,Vanlehn K. Student Assessment Using Bayesian Nets[M]. Int J of H-C S,1995.
[3]劉通江.個(gè)性化課件生成系統(tǒng)中動(dòng)態(tài)學(xué)生模型的研究[D].北京:首都師范大學(xué),2004.
[4] Conati C,Gertner A,Vanlehn K. Using Bayesian Networks to Manage Uncertainty in Student Modeling[J]. User Modeling and User-Adap? tive Instructional,2002(12).
[5]張煒,郭韶升.人工智能多媒體課件的設(shè)計(jì)與應(yīng)用[J].機(jī)械管理開(kāi)發(fā),2007,10(98):151-154 .
[6] Bunt A. On Creating a Student Model to Assess Effective Exploratory Behaviour in an Open Learning Environment[D]. Masters thesis. University of British Columbia,2001.