武善鋒 陸霞
關(guān)鍵詞: 決策樹; 數(shù)據(jù)挖掘; 體育課程; ASP.NET; 管理系統(tǒng); 課程分析
中圖分類號(hào): TN911.1?34 ? ? ? ? ? ? ? ? ? ? ? ? ? 文獻(xiàn)標(biāo)識(shí)碼: A ? ? ? ? ? ? ? ? ? ? ?文章編號(hào): 1004?373X(2019)03?0131?03
Abstract: With the continuous advancement of the information technology construction in colleges and universities, a large amount of teaching information resources are produced by teaching management of physical education curriculum. In order to improve the quality of physical education, the application scheme of decision tree algorithm in the analysis and management of physical education curriculum is proposed. The C4.5 decision tree algorithm of data mining method is analyzed. The framework and database design of the curriculum analysis and management system are given. The ASP.NET development language is used to realize the system. The SQL Server 2008 is taken as the database. Visual Studio 2010 is taken as the development environment. The test results show that the proposed physical education curriculum analysis and management system has high performance in running time and accuracy, and provides the powerful data support for improving the efficiency and quality of physical education curriculum management.
Keywords: decision tree; data mining; physical education curriculum; ASP.NET; management system; curriculum analysis
隨著計(jì)算機(jī)技術(shù)的不斷進(jìn)步和教學(xué)信息化的不斷發(fā)展,全國范圍內(nèi)的高校開始逐漸普及各種現(xiàn)代化的教學(xué)設(shè)備及相關(guān)管理系統(tǒng),如藍(lán)墨云班課、慕課(MOOC)、智慧教室互動(dòng)等[1]。信息化教學(xué)的開展和實(shí)施也產(chǎn)生了大量的各種教學(xué)管理數(shù)據(jù),針對(duì)這些大量的數(shù)據(jù),如果不去利用勢(shì)必造成巨大的資源浪費(fèi),但是如果采用人工手段去分析處理會(huì)產(chǎn)生較大的時(shí)間和人工成本,因此需要利用計(jì)算機(jī)將教師從大量的復(fù)雜和重復(fù)勞動(dòng)中解放出來[2?3]。根據(jù)不同課程的屬性和要求,提取這些數(shù)據(jù)中的必然聯(lián)系和潛在的關(guān)系已經(jīng)成為各種課程教學(xué)管理系統(tǒng)的研究方向和熱點(diǎn)。
數(shù)據(jù)挖掘作為近期世界范圍內(nèi)快速興起的一門交叉學(xué)科,匯集了來自機(jī)器學(xué)習(xí)、模式識(shí)別、數(shù)據(jù)庫、統(tǒng)計(jì)學(xué)、人工智能等各領(lǐng)域的研究成果[4?6]。計(jì)算機(jī)的大規(guī)模普及產(chǎn)生了海量的數(shù)據(jù),數(shù)據(jù)挖掘通過綜合以上學(xué)科領(lǐng)域的技術(shù)成果,對(duì)海量數(shù)據(jù)進(jìn)行處理和分析。目前,數(shù)據(jù)挖掘在教學(xué)管理系統(tǒng)中的應(yīng)用正處于初始階段,相關(guān)領(lǐng)域的研究不多,因此應(yīng)用于體育課程教學(xué)工作的案例更少,例如文獻(xiàn)[7]提出基于ID3決策樹的商務(wù)英語實(shí)踐教學(xué)成效評(píng)價(jià)方案,也就是說,現(xiàn)有的體育課程成績(jī)管理系統(tǒng)沒有成績(jī)分析功能,無法對(duì)提升體育教學(xué)工作的效率和質(zhì)量提供有力的技術(shù)支持。
3.1 ?C4.5算法在體育課程分析系統(tǒng)的應(yīng)用
以某學(xué)校20個(gè)班級(jí)的學(xué)生的體育課程成績(jī)?yōu)槔M(jìn)行數(shù)據(jù)挖掘分析,并將C4.5算法在體育課程分析系統(tǒng)中進(jìn)行具體應(yīng)用。20個(gè)班級(jí)共735個(gè)學(xué)生的訓(xùn)練集數(shù)據(jù)如表1所示。
通過表1所示的訓(xùn)練集數(shù)據(jù),運(yùn)用C4.5算法生成決策樹,程序?qū)崿F(xiàn)的部分代碼如下:
print(′Start training...′)
tree = train(train_features, train_labels, list(range(feature_len)))
time_3 = time.time()
print(′training cost %f seconds′ % (time_3 ? time_2))
print(′Start predicting...′)
test_predict = predict(test_features,tree) time_4 = time.time()
print(′predicting cost %f seconds′ % (time_4 ? time_3))
3.2 ?系統(tǒng)測(cè)試結(jié)果
對(duì)設(shè)計(jì)的體育課程分析與管理系統(tǒng)進(jìn)行功能測(cè)試和性能測(cè)試。首先,在功能測(cè)試中系統(tǒng)運(yùn)行狀態(tài)良好,操作流暢,人機(jī)交互效果良好,系統(tǒng)中學(xué)生基本信息維護(hù)界面如圖4所示。其次,在性能測(cè)試中,相比基于ID3算法的課程成績(jī)分析系統(tǒng)[7],本文系統(tǒng)處理數(shù)據(jù)所需時(shí)間減少了12%左右,同時(shí)分析數(shù)據(jù)集的準(zhǔn)確率提升了約8%。
本文提出一種基于決策樹算法的體育課程分析與管理系統(tǒng)。首先對(duì)成績(jī)分析的需求進(jìn)行研究,并給出課程分析與管理系統(tǒng)的框架及其數(shù)據(jù)庫設(shè)計(jì)。然后采用優(yōu)化后的C4.5決策樹算法實(shí)現(xiàn)具體數(shù)據(jù)挖掘。采用ASP.NET開發(fā)語言,數(shù)據(jù)庫為SQL Server 2008,開發(fā)環(huán)境為Visual Studio 2010。該系統(tǒng)利用決策樹算法提取體育課程工作中的數(shù)據(jù)特征和關(guān)系,并結(jié)合成績(jī)分析形成可參考的學(xué)生個(gè)性化信息,為體育課程的教學(xué)管理提供了有價(jià)值的數(shù)據(jù)支持和理論參考,該系統(tǒng)可有效提高體育課程教學(xué)和管理的質(zhì)量和效率。
參考文獻(xiàn)
[1] 張怡婷,付雄,陳蕾.算法課程面向應(yīng)用的教學(xué)模式研究[J].計(jì)算機(jī)教育,2012(24):82?86.
ZHANG Yiting, FU Xiong, CHEN Lei. Research on application?oriented teaching model of algorithm course [J]. Computer education, 2012(24): 82?86.
[2] 茹峰,彭曉宏,侯立剛,等.基于人臉識(shí)別的學(xué)生課程出勤管理系統(tǒng)設(shè)計(jì)[J].現(xiàn)代電子技術(shù),2016,39(1):81?84.
RU Feng, PENG Xiaohong, HOU Ligang, et al. Design of student course attendance management system based on face re?cognition [J]. Modern electronics technique, 2016, 39(1): 81?84.
[3] 鄧?yán)颦偅で锔?,李中寧,?基于大數(shù)據(jù)分析的計(jì)算機(jī)網(wǎng)絡(luò)課程自適應(yīng)學(xué)習(xí)系統(tǒng)構(gòu)建[J].計(jì)算機(jī)工程與科學(xué),2016,38(z1):276?280.
DENG Liqiong, XIAO Qiugen, LI Zhongning, et al. Construction of adaptive learning system for computer network course based on big data analysis [J]. Computer engineering and science, 2016, 38(S1): 276?280.
[4] JOSEPH S R, HLOMANI H, LETSHOLO K. Data mining algorithms: an overview [J]. Neuroscience, 2016, 12(3): 719?743.
[5] NGAI E W T, XIU L, CHAU D C K. Application of data mi?ning techniques in customer relationship management: a literature review and classification [J]. Expert systems with applications, 2012, 36(2): 2592?2602.
[6] HARPAZ R, DUMOUCHEL W, SHAH N H, et al. Novel data mining methodologies for adverse drug event discovery and analysis [J]. Clinical pharmacology & therapeutics, 2012, 91(6): 1010?1021.
[7] 余洪英.基于決策樹的商務(wù)英語實(shí)踐教學(xué)成效評(píng)價(jià)[J].科技通報(bào),2016,32(2):229?232.
YU Hongying. Evaluation of the effectiveness of business English practice teaching based on decision tree [J]. Bulletin of science and technology, 2016, 32(2): 229?232.
[8] SAFAVIAN S R, LANDGREBE D. A survey of decision tree classifier methodology [J]. IEEE transactions on systems, man & cybernetics, 2002, 21(3): 660?674.
[9] GLIGOROV V V, WILLIAMS M. Efficient, reliable and fast high?level triggering using a bonsai boosted decision tree [J]. Journal of instrumentation, 2012, 8(2):6.
[10] BARROS R C, FREITAS A A. A survey of evolutionary algorithms for decision?tree induction [J]. IEEE transactions on systems, man & cybernetics Part C, 2012, 42(3): 291?312.