个性化在线答疑系统的研究与实现
|
摘要
随着计算机网络技术和多媒体技术的发展,远程教育越来越受到人们的重视。答疑系统作为远程教育系统中的一个重要组成部分,直接关系着整个系统的教学效果,因而越来越受到教学人员和远程教育系统开发人员的关注。
本文在综述国内外答疑系统研究现状的基础上,分析了目前答疑系统存在的主要局限:(1)由于现有智能技术还不能实现对自然语言的完全理解,所以基于自然语言理解的答疑系统的智能化程度十分有限;(2)没有充分利用概念之间的语义关系来引导学习者进一步的提问;(3)缺少对学习者疑问特征的分析,不能主动为学习者推送其存在疑问或感兴趣的知识;(4)没有发挥网络的优势,为答疑系统的问题库提供丰富的参考资源。
针对现有答疑系统的上述缺陷,本文设计并实现了一个特定领域的具有个性化服务能力的、基于自然语言理解的在线答疑系统。本论文的主要工作包括:(1)分析国内外网络答疑系统的现状,以及当前自然语言处理和个性化服务的相关技术;(2)用主题爬虫技术为答疑问题库采集问题,不断丰富问题库;(3)开发了一个具有个性化服务能力的、可以主动为学习者推送信息的、基于自然语言理解的答疑系统;并以浙江师范大学国家精品课程《现代教育技术》为例对所开发的答疑系统进行了应用。实验证明,该个性化的答疑系统可以为学习者推送其感兴趣的内容,促进学习者对问题相关内容的理解,开阔学习者的思路。
With the development of the computer network technology and multi-media technology, distance education has been paid more and more attention to. As an important component of distance education system, question answering system exerts direct influence on the quality of distance education, thus being attached more and more importance to by teachers and developers of distance education system.
Based on the comprehensive narration of the internal and external current research situations, this dissertation pointed to the major limitations of present question answering system in China: 1. the intelligence subjected to the natural language understanding is rather limited, for the artificial intelligence cannot make the total understanding of natural language come true.; 2. It does not make use of the semantic relations of concepts to guide the users to ask further questions; 3. It cannot positively show learners the knowledge they feel puzzled at and interested in because of lacking the analysis of the symptoms of learners' confusion; 4. It does not supply abundant reference resources to the question database with the advantage of internet.
Aiming at the remedy for the above limitations of present question answering system in China, this dissertation develops an on-line question answering system based on natural language understanding, which provides personalized service. The major tasks of this research include: 1. the analysis of current situations of the internal and external answer systems, and of the technologies related to the present personalized service and natural language understanding; 2. collection of questions through the focused crawler with the purpose of continuously enriching the question database; 3. developing an question answering system with "personalized service, positive sending of information to learners and basis on natural language understanding, which has been applied into the national excellent course "Modern Education Technology" developed by Zhejiang Normal University, as an example. The experiment proves that this personalized question answering system can send the content learners fell interested in, improve learners' comprehension of content related to the questions, and open the thoughts of learners.
本文在综述国内外答疑系统研究现状的基础上,分析了目前答疑系统存在的主要局限:(1)由于现有智能技术还不能实现对自然语言的完全理解,所以基于自然语言理解的答疑系统的智能化程度十分有限;(2)没有充分利用概念之间的语义关系来引导学习者进一步的提问;(3)缺少对学习者疑问特征的分析,不能主动为学习者推送其存在疑问或感兴趣的知识;(4)没有发挥网络的优势,为答疑系统的问题库提供丰富的参考资源。
针对现有答疑系统的上述缺陷,本文设计并实现了一个特定领域的具有个性化服务能力的、基于自然语言理解的在线答疑系统。本论文的主要工作包括:(1)分析国内外网络答疑系统的现状,以及当前自然语言处理和个性化服务的相关技术;(2)用主题爬虫技术为答疑问题库采集问题,不断丰富问题库;(3)开发了一个具有个性化服务能力的、可以主动为学习者推送信息的、基于自然语言理解的答疑系统;并以浙江师范大学国家精品课程《现代教育技术》为例对所开发的答疑系统进行了应用。实验证明,该个性化的答疑系统可以为学习者推送其感兴趣的内容,促进学习者对问题相关内容的理解,开阔学习者的思路。
With the development of the computer network technology and multi-media technology, distance education has been paid more and more attention to. As an important component of distance education system, question answering system exerts direct influence on the quality of distance education, thus being attached more and more importance to by teachers and developers of distance education system.
Based on the comprehensive narration of the internal and external current research situations, this dissertation pointed to the major limitations of present question answering system in China: 1. the intelligence subjected to the natural language understanding is rather limited, for the artificial intelligence cannot make the total understanding of natural language come true.; 2. It does not make use of the semantic relations of concepts to guide the users to ask further questions; 3. It cannot positively show learners the knowledge they feel puzzled at and interested in because of lacking the analysis of the symptoms of learners' confusion; 4. It does not supply abundant reference resources to the question database with the advantage of internet.
Aiming at the remedy for the above limitations of present question answering system in China, this dissertation develops an on-line question answering system based on natural language understanding, which provides personalized service. The major tasks of this research include: 1. the analysis of current situations of the internal and external answer systems, and of the technologies related to the present personalized service and natural language understanding; 2. collection of questions through the focused crawler with the purpose of continuously enriching the question database; 3. developing an question answering system with "personalized service, positive sending of information to learners and basis on natural language understanding, which has been applied into the national excellent course "Modern Education Technology" developed by Zhejiang Normal University, as an example. The experiment proves that this personalized question answering system can send the content learners fell interested in, improve learners' comprehension of content related to the questions, and open the thoughts of learners.
引文
铩颷1]郑庆华,胡云华,张素娟.自然语言网络答疑系统的研究与实现[J].小型微型计算机系统,2005,26(3):554-555
[2]柳泉波,黄荣怀,何克抗.智能答疑系统的设计与实现[J].中国远程教育,2000,(8):43-44
[3]Ask jeeves for kids[DB/OL].http://www.askforkids.com/,2007-1-8
[4]AnswerBus[DB/OL].http://answerbus.coli.uni-saarland.de/index.shtml,2007-1-8
[5]START[DB/OL],Natural Language Question AnsweringSystem.http://start.csail.mit.edu/,2007-1-8
[6]李爽,陈丽.国内外网上智能答疑系统比较研究[J].中国电化教育,2003,(1):80-82
[7]周文帅,冯速.汉语分词技术研究现状与应用展望[J].山西师范大学学报(自然科学版),2006,20(1):27-28
[8]文庭孝,邱均平,侯经川.汉语自动分词研究展望[J].现代图书情报技术,2004(7):6-10
[9]沈斌.基于分词的中文文本相似度计算研究[D].天津财经大学,2006:6
[10]彭喜化,黄睿.基于向量空间模型的多关键字检索技术[J].重庆职业技术学院学报,2006,15(3):151
[11]董振东,董强.知网[DB/OL].http://www.keenage.com,2007-1-11
[12]李志辉.基于专业知网的中文智能答疑系统模型研究[D].西南大学,2006:14-15
[13]张同珍,申瑞民.基于Web的自动答疑系统问题匹配算法研究与实现[J].计算机工程与应用,2003.29:104
[14]李志辉.答疑系统中概念词典的设计与应用[J].重庆科技学院学报(社会科学版),2005,2:87-88
[15]曾春,邢春晓,周立柱.个性化服务技术综述[J].软件学报,2002,13(10):1953-1960
[16]Claypool,M.,Le,P.,Waseda,M.,et al.Implicit interest indicators.In:Campbell,M.,ed.Proceedings of the ACM Intelligent User Interfaces Conference(IUI).New York:ACM Press,2001.14-17.
[17]刘洁清.网站聚焦爬虫研究[D].江西财经大学,2006:8-9
[18]汪涛,樊孝忠.主题爬虫的设计与实现[J].计算机应用,2004,(6):270-271
[19]夏诏杰,梁春燕,郭力.化学主题网络爬虫的设计和实现[J].计算机工程与应用,2006,(10):204
[20]王巍.基于的汉语自动分词算法研究[J].中州大学学报,2007,24(1):120
[21]刘玮玮.搜索引擎中主题爬虫的研究与实现[D].南京理工大学,2006:25、38
[22]蒋科.基于领域概念定制的主题爬虫系统的设计与实现[D].西安电子科技大学,2007:22
铩颷23]孙承杰,关毅.基于统计的网页正文信息抽取方法的研究[J].中文信息学报,2004,18(5):17-22
[24].NET的优点[DB/OL].http://zhidao.baidu.com/question/5715764.html,2007-6-1
[25].NET的优点[DB/OL].http://www.faq120.com/bbs/read.php?tid=435,2007-6-1
[26].NET的优点[DB/OL].http://msdn2.microsoft.com/zh-cn/library/default.aspx,2007-6-1
[27]单林娜.基于.NET的网上书店的设计与实现[D].重庆大学,2005:9
[28]程建钢等.基于Web的远程答疑系统[J].电化教育研究,2002,7:56
[29]刘忠,王成道.基于汉语自然语言信息查询的计算机理解实现[J].计算机应用,2004,24(1):8-11
[30]洪小伟,刘亚军.基于XML智能答疑系统的知识库构建[J].微机发展,2003,13(8):1-2
[31]赵赟,刘亚军.智能答疑系统中语义网的研究与应用[J].微机发展,2003,13(11):36-37、40
[32]汪红兵,佘春东,范植华等.基于JMS的数据推送系统的设计与实现[J].计算机应用,2005,25(12):366-368
[33]张晓辉,何丕廉,刘光然.智能答疑系统中搜索技术的研究[J].微计算机应用,2006,27(3):261-263
[34]徐科,崔志明.基于搜索历史的用户兴趣模型的研究[J].计算机技术与发展,2006,16(5)-18-20
[35]欧洁.基于贝叶斯网络模型的用户兴趣联合推送[J].计算机科学,2003,30(12):73-77
[36]何波,王越.基于数据挖掘的Web个性化信息推荐系统[J].计算机工程与应用,2006,3:178-179、186
[37]汪涛,樊孝忠,顾益军等.基于概念分析的主题爬虫设计[J].北京理工大学学报,2004,24(10):890-893
[38]周立柱,林玲聚.焦爬虫技术研究综述[J].计算机应用,2005,25(9):1965-1969
[39]Schafer,J.B.,Konstan,J.A,and Riedl,J.E-Commerce Recommendation Applications.Data Mining and Knowledge Discovery.2001:65-69
[40]Karypis,G.Evaluation of Item-based top-n Recommendation Algorithms.InProceedings of the Tenth International Conference on Information andKnowledge Management(CIKM).2001:267-274
[2]柳泉波,黄荣怀,何克抗.智能答疑系统的设计与实现[J].中国远程教育,2000,(8):43-44
[3]Ask jeeves for kids[DB/OL].http://www.askforkids.com/,2007-1-8
[4]AnswerBus[DB/OL].http://answerbus.coli.uni-saarland.de/index.shtml,2007-1-8
[5]START[DB/OL],Natural Language Question AnsweringSystem.http://start.csail.mit.edu/,2007-1-8
[6]李爽,陈丽.国内外网上智能答疑系统比较研究[J].中国电化教育,2003,(1):80-82
[7]周文帅,冯速.汉语分词技术研究现状与应用展望[J].山西师范大学学报(自然科学版),2006,20(1):27-28
[8]文庭孝,邱均平,侯经川.汉语自动分词研究展望[J].现代图书情报技术,2004(7):6-10
[9]沈斌.基于分词的中文文本相似度计算研究[D].天津财经大学,2006:6
[10]彭喜化,黄睿.基于向量空间模型的多关键字检索技术[J].重庆职业技术学院学报,2006,15(3):151
[11]董振东,董强.知网[DB/OL].http://www.keenage.com,2007-1-11
[12]李志辉.基于专业知网的中文智能答疑系统模型研究[D].西南大学,2006:14-15
[13]张同珍,申瑞民.基于Web的自动答疑系统问题匹配算法研究与实现[J].计算机工程与应用,2003.29:104
[14]李志辉.答疑系统中概念词典的设计与应用[J].重庆科技学院学报(社会科学版),2005,2:87-88
[15]曾春,邢春晓,周立柱.个性化服务技术综述[J].软件学报,2002,13(10):1953-1960
[16]Claypool,M.,Le,P.,Waseda,M.,et al.Implicit interest indicators.In:Campbell,M.,ed.Proceedings of the ACM Intelligent User Interfaces Conference(IUI).New York:ACM Press,2001.14-17.
[17]刘洁清.网站聚焦爬虫研究[D].江西财经大学,2006:8-9
[18]汪涛,樊孝忠.主题爬虫的设计与实现[J].计算机应用,2004,(6):270-271
[19]夏诏杰,梁春燕,郭力.化学主题网络爬虫的设计和实现[J].计算机工程与应用,2006,(10):204
[20]王巍.基于的汉语自动分词算法研究[J].中州大学学报,2007,24(1):120
[21]刘玮玮.搜索引擎中主题爬虫的研究与实现[D].南京理工大学,2006:25、38
[22]蒋科.基于领域概念定制的主题爬虫系统的设计与实现[D].西安电子科技大学,2007:22
铩颷23]孙承杰,关毅.基于统计的网页正文信息抽取方法的研究[J].中文信息学报,2004,18(5):17-22
[24].NET的优点[DB/OL].http://zhidao.baidu.com/question/5715764.html,2007-6-1
[25].NET的优点[DB/OL].http://www.faq120.com/bbs/read.php?tid=435,2007-6-1
[26].NET的优点[DB/OL].http://msdn2.microsoft.com/zh-cn/library/default.aspx,2007-6-1
[27]单林娜.基于.NET的网上书店的设计与实现[D].重庆大学,2005:9
[28]程建钢等.基于Web的远程答疑系统[J].电化教育研究,2002,7:56
[29]刘忠,王成道.基于汉语自然语言信息查询的计算机理解实现[J].计算机应用,2004,24(1):8-11
[30]洪小伟,刘亚军.基于XML智能答疑系统的知识库构建[J].微机发展,2003,13(8):1-2
[31]赵赟,刘亚军.智能答疑系统中语义网的研究与应用[J].微机发展,2003,13(11):36-37、40
[32]汪红兵,佘春东,范植华等.基于JMS的数据推送系统的设计与实现[J].计算机应用,2005,25(12):366-368
[33]张晓辉,何丕廉,刘光然.智能答疑系统中搜索技术的研究[J].微计算机应用,2006,27(3):261-263
[34]徐科,崔志明.基于搜索历史的用户兴趣模型的研究[J].计算机技术与发展,2006,16(5)-18-20
[35]欧洁.基于贝叶斯网络模型的用户兴趣联合推送[J].计算机科学,2003,30(12):73-77
[36]何波,王越.基于数据挖掘的Web个性化信息推荐系统[J].计算机工程与应用,2006,3:178-179、186
[37]汪涛,樊孝忠,顾益军等.基于概念分析的主题爬虫设计[J].北京理工大学学报,2004,24(10):890-893
[38]周立柱,林玲聚.焦爬虫技术研究综述[J].计算机应用,2005,25(9):1965-1969
[39]Schafer,J.B.,Konstan,J.A,and Riedl,J.E-Commerce Recommendation Applications.Data Mining and Knowledge Discovery.2001:65-69
[40]Karypis,G.Evaluation of Item-based top-n Recommendation Algorithms.InProceedings of the Tenth International Conference on Information andKnowledge Management(CIKM).2001:267-274