基于网络教学平台的试题库组卷算法研究
|
摘要
网络教学平台主要针对高校的教学与学习提供全面服务的网络支撑平台。试题库系统是网络教学平台的重要组成部分,而自动组卷是试题库系统的核心。本论文基于昆明理工大学计算中心教学改革项目——计算机基础教学网络教学平台的建设与应用,对其中试题库组卷算法进行深入的研究。
首先,研究并确定本试题库采用的理论:经典测量理论CTT;同时阐述了基于经典测量理论的项目分析指标体系和评价标准。
其次,研究试题库的数据组织结构,确立本试题库数据的组织方式并对试题库中数据表的结构进行了分析和设计。
然后,分析探讨现有的几种组卷算法,结合本教学平台试题库的结构及其中试题数量不庞大的特点,采用改进的随机算法来实现试题库的自动组卷,实践结果表明该组卷算法结构简单、实现起来也简单而且非常实用,组卷速度、组卷质量已达到所需要求。
最后,针对当试题库中题量增大以及组卷约束条件增多时,改进的随机算法可能难以满足要求的问题,结合本试题库的结构特点,尝试性地研究了一种基于模块的遗传组卷算法,该算法使用了模块的思想,把相同题型的试题放在同一模块,在进行交叉、变异时,保持在模块内部,对今后组卷系统的改进有一定的借鉴作用。
The Network teaching platform is a Network supporting platform, providing the all-around service mainly aiming at the teaching and study of the institutions of higher learning. The bank of test questions is the important part of the platform, and the core is the test paper auto-generation. This paper lucubrated the generation algorithms of the bank of test questions, basing on the building and application of the Netword teaching platform of the computer basic teaching, which is the teaching reformation item of the Computer Center in the Kunming University of Science and Technology.
Firstly, we researched and confirmed the theories of this bank of test questions:the Classic Test Theories(CTT),and expatiated the item analyzed system and evaluating standard based on CTT.
Secondly, we studied the structure of organizing data in the bank of test questions, then we established the fashion of organizing data in this bank of test questions and analyzed and designed the datasheet structure.
Thirdly,we analyzed and discussed several common generation algorithms, then we adopted the advanced random algorithm to realizing the test paper auto-generation of this bank of test questions, considering the database structure of this teaching platform and the number of questions is not enormous. The practised result showed that this algorithm' s structure is simple and realized easily, and it is practical very much, the speed and quality of the test paper generation already arrived at the request.
In the end, if the number of questions in this bank of test questions and the constraint conditions of test paper generation increase, this advanced random algorithm can' t meet our needs, then thinking over the structure characteristic of this database, we tried to study a advanced genetic algorithm basing on the module, which used module' s idea, namely the same-type questions are put together, then the crossing and mutating are in the module. This genetic algorithm has the use for reference to the future test paper generation system' s improvement.
首先,研究并确定本试题库采用的理论:经典测量理论CTT;同时阐述了基于经典测量理论的项目分析指标体系和评价标准。
其次,研究试题库的数据组织结构,确立本试题库数据的组织方式并对试题库中数据表的结构进行了分析和设计。
然后,分析探讨现有的几种组卷算法,结合本教学平台试题库的结构及其中试题数量不庞大的特点,采用改进的随机算法来实现试题库的自动组卷,实践结果表明该组卷算法结构简单、实现起来也简单而且非常实用,组卷速度、组卷质量已达到所需要求。
最后,针对当试题库中题量增大以及组卷约束条件增多时,改进的随机算法可能难以满足要求的问题,结合本试题库的结构特点,尝试性地研究了一种基于模块的遗传组卷算法,该算法使用了模块的思想,把相同题型的试题放在同一模块,在进行交叉、变异时,保持在模块内部,对今后组卷系统的改进有一定的借鉴作用。
The Network teaching platform is a Network supporting platform, providing the all-around service mainly aiming at the teaching and study of the institutions of higher learning. The bank of test questions is the important part of the platform, and the core is the test paper auto-generation. This paper lucubrated the generation algorithms of the bank of test questions, basing on the building and application of the Netword teaching platform of the computer basic teaching, which is the teaching reformation item of the Computer Center in the Kunming University of Science and Technology.
Firstly, we researched and confirmed the theories of this bank of test questions:the Classic Test Theories(CTT),and expatiated the item analyzed system and evaluating standard based on CTT.
Secondly, we studied the structure of organizing data in the bank of test questions, then we established the fashion of organizing data in this bank of test questions and analyzed and designed the datasheet structure.
Thirdly,we analyzed and discussed several common generation algorithms, then we adopted the advanced random algorithm to realizing the test paper auto-generation of this bank of test questions, considering the database structure of this teaching platform and the number of questions is not enormous. The practised result showed that this algorithm' s structure is simple and realized easily, and it is practical very much, the speed and quality of the test paper generation already arrived at the request.
In the end, if the number of questions in this bank of test questions and the constraint conditions of test paper generation increase, this advanced random algorithm can' t meet our needs, then thinking over the structure characteristic of this database, we tried to study a advanced genetic algorithm basing on the module, which used module' s idea, namely the same-type questions are put together, then the crossing and mutating are in the module. This genetic algorithm has the use for reference to the future test paper generation system' s improvement.
引文
[1]何克抗.建立题库理论.长沙:国防科技大学出版社,1995,52-117.
[2]谢深泉,胡宁静.数据库设计和自动组卷中的几个问题.湘潭大学自然科学学报.2002,24(3):27-31
[3]刘树刚.一种Web测试中的组卷策略及其改进.烟台大学学报.2004,17(1):59-64
[4]李新国.基于VF6.0的学校考试自动组卷系统.计算机工程与设计.2003.24(11):66-69
[5]彭勇.基于区分度的智能组卷难度正态分布算法.微机发展.2003,13(11):45-47
[6]刘彬等.智能组卷系统试题库结构的研究.信息技术.2002,3(7):2-4
[7]熊伟清,魏平.一个求解组卷问题的遗传算法设计.计算机应用与软件.2003,9(3):69-71
[8]毛秉毅.基于遗传算法的智能组卷系统数据库结构的研究.计算机工程与应用.2003,39(6):230-232
[9]孙勇,柏云.基于遗传算法的试题组卷策略.淄博学院学报.2002,4(3):27-28
[10]石中盘,韩卫.基于概率论和自适应遗传算法的智能抽题算法.计算机工程.2002,28(1):141-143
[11]董敏,霍剑青,王晓蒲.基于自适应遗传算法的智能组卷研究.小型微型计算机系统.2004,25(1):82-85
[12]Win J.van der Linden,Bernard P.etal.An integer programming approach to item bank design[J].Applied Psychological Measurement.2000,24(2):139-150
[13]Huub Verstralen,Timo Bechger,Gunter Maris.The combined use of classical test and item response theory,http://www.cito.ni/pok/eind_fr.htm
[14]De Jong K,Spears W.Using genetic algorithms to solve NP-completep roblems.Proceedings of the Third International Conference on Genetic A gorithms,1989
[15]HMannila,HToivonen,InkeriVerkamo.A Efficient algorithm for discovering Association rulers.In Processing of AAAI Workshop on Knowledge Discovering in Database.1994:181-192
[16]R Srikant,R Agrawal.Mining generalized association rulers.Proceedings of the 21th international Conference on very Large.Dattabases.1995:407-419
[17]R.Agrawal,T Imielinski,Aswami.Mining Associations Rules between Sets of Items in Large Databases,Proc,ACMSIGMOD.1993:207-216
[18]R.Agrawal,R.Srikant.Fast Algorithms for MiningAssociation Rules in Large Databases.Proc.2Oth IntConf.1994:478-499
[19]J Spark.An Effective Hash based Algorithm for Mining Association rules.Proc.ACM.SIGMOD.1995:175-186
[20]Hannu toivonen.Sampling larged atabases for association rules,ln Proceedings of the 22th international conference on Very Large Databases.Bombay,India MorganKaufmann.1996:134-145
[21]J R Quinlan.Induction of decision trees.Machine Learning.1986,(1):81-106
[22]S.Chaudhuri and U.Dayal.An review of datawarehousing and OLAP technology.SIGMOD Record.1996,(26):65-74
[23]Piatetsky,Shapiro.Knowledge Discovery in Databases.AAAI/MIT Press.1991:20-23
[24]J R Quinlan.Induction of induction of decision tress.Machine learning.1986,(1):81-106
[25]W Buntine,T Niblett.A further comparison of splitting rules for decision-tree induction.Machine Learning.1992,8:75-8
[26]王耀南.智能控制系统——模糊逻辑、专家系统、神经网络.长沙:湖南大学出版社.1996:32-48
[27]汪德顺,顾学春,马立兴.基于知识的成卷方法及在MATBAS系统中的实现.西安交大学报.1988,23(1):35-39
[28]熊伟清,胡军.一种题库模型与组卷算法.兰州铁道学院学报.1999,18(2):85-88
[29]叶勇,刘峰.智能组卷中提高系统运行效率的算法研究微机发展.1998,4(1)14-17
[30]谢平.基于框架模式的试题库智能组卷系统.华东交通大学学报.1998,15(4):58-63
[31]全惠云,范国闯,赵霆雷.基于遗传算法的试题库智能组卷系统研究.武汉大学学报(自然科学版).1999,45(5):758-760
[32]王学兰编著.教育统计学.南开大学出版社,1997年6月.
[33]陈丽燕.试题库系统数据组织结构的探讨.丽水学院学报,2004,26.
[34]林雪明,张钧良.基于知识点的试题库组卷算法的建立[J].微机发展,2001年2月:77-79.
[35]刘贤喜,杨峰等.基于遗传算法的自动组卷的研究.科学技术与工程,2007.5.
[36]王小平,曹立明著.遗传算法——理论,应用与软件实现.西安交通大学出版社,2002.1.
[37]余胜泉,何克抗.网络教学平台的体系结构与功能.中国电化教育,2001.8.
[38]李敏强等著.遗传算法的基本理论与应用.科学出版社,2002.3.
[39]陈丽娜.基于遗传算法的试题库组卷方法研究.华东师范大学硕士论文,2005.10.
[40]万明秀.基于网络教学平台的考试系统的设计与实现,天津师范大学硕士学位论文,2005.4.
[41]曹轶群,孙一江等编著.PHP高级开发技术与应用.清华大学出版社,2002.5.
[42]M.H.Al suwaiyel著,吴伟昶等译.算法设计技巧与分析.电子工业出版社,2004.8.
[2]谢深泉,胡宁静.数据库设计和自动组卷中的几个问题.湘潭大学自然科学学报.2002,24(3):27-31
[3]刘树刚.一种Web测试中的组卷策略及其改进.烟台大学学报.2004,17(1):59-64
[4]李新国.基于VF6.0的学校考试自动组卷系统.计算机工程与设计.2003.24(11):66-69
[5]彭勇.基于区分度的智能组卷难度正态分布算法.微机发展.2003,13(11):45-47
[6]刘彬等.智能组卷系统试题库结构的研究.信息技术.2002,3(7):2-4
[7]熊伟清,魏平.一个求解组卷问题的遗传算法设计.计算机应用与软件.2003,9(3):69-71
[8]毛秉毅.基于遗传算法的智能组卷系统数据库结构的研究.计算机工程与应用.2003,39(6):230-232
[9]孙勇,柏云.基于遗传算法的试题组卷策略.淄博学院学报.2002,4(3):27-28
[10]石中盘,韩卫.基于概率论和自适应遗传算法的智能抽题算法.计算机工程.2002,28(1):141-143
[11]董敏,霍剑青,王晓蒲.基于自适应遗传算法的智能组卷研究.小型微型计算机系统.2004,25(1):82-85
[12]Win J.van der Linden,Bernard P.etal.An integer programming approach to item bank design[J].Applied Psychological Measurement.2000,24(2):139-150
[13]Huub Verstralen,Timo Bechger,Gunter Maris.The combined use of classical test and item response theory,http://www.cito.ni/pok/eind_fr.htm
[14]De Jong K,Spears W.Using genetic algorithms to solve NP-completep roblems.Proceedings of the Third International Conference on Genetic A gorithms,1989
[15]HMannila,HToivonen,InkeriVerkamo.A Efficient algorithm for discovering Association rulers.In Processing of AAAI Workshop on Knowledge Discovering in Database.1994:181-192
[16]R Srikant,R Agrawal.Mining generalized association rulers.Proceedings of the 21th international Conference on very Large.Dattabases.1995:407-419
[17]R.Agrawal,T Imielinski,Aswami.Mining Associations Rules between Sets of Items in Large Databases,Proc,ACMSIGMOD.1993:207-216
[18]R.Agrawal,R.Srikant.Fast Algorithms for MiningAssociation Rules in Large Databases.Proc.2Oth IntConf.1994:478-499
[19]J Spark.An Effective Hash based Algorithm for Mining Association rules.Proc.ACM.SIGMOD.1995:175-186
[20]Hannu toivonen.Sampling larged atabases for association rules,ln Proceedings of the 22th international conference on Very Large Databases.Bombay,India MorganKaufmann.1996:134-145
[21]J R Quinlan.Induction of decision trees.Machine Learning.1986,(1):81-106
[22]S.Chaudhuri and U.Dayal.An review of datawarehousing and OLAP technology.SIGMOD Record.1996,(26):65-74
[23]Piatetsky,Shapiro.Knowledge Discovery in Databases.AAAI/MIT Press.1991:20-23
[24]J R Quinlan.Induction of induction of decision tress.Machine learning.1986,(1):81-106
[25]W Buntine,T Niblett.A further comparison of splitting rules for decision-tree induction.Machine Learning.1992,8:75-8
[26]王耀南.智能控制系统——模糊逻辑、专家系统、神经网络.长沙:湖南大学出版社.1996:32-48
[27]汪德顺,顾学春,马立兴.基于知识的成卷方法及在MATBAS系统中的实现.西安交大学报.1988,23(1):35-39
[28]熊伟清,胡军.一种题库模型与组卷算法.兰州铁道学院学报.1999,18(2):85-88
[29]叶勇,刘峰.智能组卷中提高系统运行效率的算法研究微机发展.1998,4(1)14-17
[30]谢平.基于框架模式的试题库智能组卷系统.华东交通大学学报.1998,15(4):58-63
[31]全惠云,范国闯,赵霆雷.基于遗传算法的试题库智能组卷系统研究.武汉大学学报(自然科学版).1999,45(5):758-760
[32]王学兰编著.教育统计学.南开大学出版社,1997年6月.
[33]陈丽燕.试题库系统数据组织结构的探讨.丽水学院学报,2004,26.
[34]林雪明,张钧良.基于知识点的试题库组卷算法的建立[J].微机发展,2001年2月:77-79.
[35]刘贤喜,杨峰等.基于遗传算法的自动组卷的研究.科学技术与工程,2007.5.
[36]王小平,曹立明著.遗传算法——理论,应用与软件实现.西安交通大学出版社,2002.1.
[37]余胜泉,何克抗.网络教学平台的体系结构与功能.中国电化教育,2001.8.
[38]李敏强等著.遗传算法的基本理论与应用.科学出版社,2002.3.
[39]陈丽娜.基于遗传算法的试题库组卷方法研究.华东师范大学硕士论文,2005.10.
[40]万明秀.基于网络教学平台的考试系统的设计与实现,天津师范大学硕士学位论文,2005.4.
[41]曹轶群,孙一江等编著.PHP高级开发技术与应用.清华大学出版社,2002.5.
[42]M.H.Al suwaiyel著,吴伟昶等译.算法设计技巧与分析.电子工业出版社,2004.8.