远程教学中自动答疑系统的研究
|
摘要
随着Internet在我国广泛应用，远程教育越来越受到人们的重视。远程教学不仅仅是将教学材料在网上发布，更多的是学生与教师、教师与学生之间的充分沟通、交流。由于远程教学中教师与学生之间在空间上是分离的，沟通与交流就显得尤为重要。因此，答疑系统是远程教学的重要组成部分。
在传统课堂教学模式中，答疑是通过师生面对面方式进行的，远程教育中这种答疑方式已经不再可行。目前远程教学系统主要采用电子邮件、在线讨论和关键字查询方式三种主要方式进行答疑。对于电子邮件方式，教师不能及时地把答案反馈给学生，同一问题多次回答；在线讨论方式要求教师必须时时在线。这两种方式都造成了教师资源和答案资源的巨大浪费。关键词查询方式要求用户具备一定的关键词抽取技术，给用户增加负担，查询效果并不理想。要解决以上问题，必须引入新技术，对现有的答疑系统进行改进。
作者利用检索系统中自动问答系统的工作原理，采用自然语言处理技术，根据教学答疑系统的特点，提出了一种在限定领域中支持自然语言理解的自动答疑系统模型，并对其关键技术和辅助手段进行分析和讨论。该系统能够自动理解用户输入的自然语言问题文本，返回一组与问题相关的答案，具有节约资源、智能性、执行效率高等特点。
本文首先对国内外答疑系统进行了对比，指出我国答疑系统缺少先进技术支持，接着阐述自然语言处理技术原理，给出自动答疑系统总体设计方案，然后对系统进行实现，并对系统关键技术进行讨论和改进，最后进行总结，指出该类型答疑系统具有广阔的研究前景和实用价值。
本文的主要工作如下：
●设计词语权重的计算方法，并给出实现方案；
●句子相似度算法的改进；
●索引和二级检索策略的引用。
With the development of Internet in our country, the people pay more attention to the distance education. The distance education not only means putting the study information on the net, but also means communicating between students and teachers. Communication is more importance, because of the distance. So, the question answer system is one key parts of the distance education.
The face to face can't appear in the distance education that always is seen in the traditional education. There three main question answer systems: e-mail, charting and key word searching. The shortcoming of e-mail is that this system can't give students answer in time and answer several times for the same question; Charting asks teachers on the net all the time. They all waste a lot of teacher's and answer's resource. The third of system needs students know much knowledge about the searching and increases students burden. The article imports some new technology and improves the model of question answer system to resolve those questions.
The author makes use of natural language process technology and advances a new sort of question answer system of natural language in limited domain. The system accept the natural language question and know what about the question, then the system give out a set of answers related to the question. There are three Characteristics in the question answer system, retrenching resource, intelligence, and high efficiency
Comparing with the foreign question answer system, the paper points out that the shortcoming of our system is lack of advancing technology. Firstly, the article introduces the technology about natural language process and designs the structure of the system. Secondly, the author turns up this system and discusses some key art and makes some program on the arithmetic. Finally, the paper thinks that the system is utility and is deeply researched in the future.
The main work of this paper is as the following:
Designing precept of the key of and realizing it;
Improving arithmetic of similarity;
Importing index structure and the tactic of the two level searching.
在传统课堂教学模式中，答疑是通过师生面对面方式进行的，远程教育中这种答疑方式已经不再可行。目前远程教学系统主要采用电子邮件、在线讨论和关键字查询方式三种主要方式进行答疑。对于电子邮件方式，教师不能及时地把答案反馈给学生，同一问题多次回答；在线讨论方式要求教师必须时时在线。这两种方式都造成了教师资源和答案资源的巨大浪费。关键词查询方式要求用户具备一定的关键词抽取技术，给用户增加负担，查询效果并不理想。要解决以上问题，必须引入新技术，对现有的答疑系统进行改进。
作者利用检索系统中自动问答系统的工作原理，采用自然语言处理技术，根据教学答疑系统的特点，提出了一种在限定领域中支持自然语言理解的自动答疑系统模型，并对其关键技术和辅助手段进行分析和讨论。该系统能够自动理解用户输入的自然语言问题文本，返回一组与问题相关的答案，具有节约资源、智能性、执行效率高等特点。
本文首先对国内外答疑系统进行了对比，指出我国答疑系统缺少先进技术支持，接着阐述自然语言处理技术原理，给出自动答疑系统总体设计方案，然后对系统进行实现，并对系统关键技术进行讨论和改进，最后进行总结，指出该类型答疑系统具有广阔的研究前景和实用价值。
本文的主要工作如下：
●设计词语权重的计算方法，并给出实现方案；
●句子相似度算法的改进；
●索引和二级检索策略的引用。
With the development of Internet in our country, the people pay more attention to the distance education. The distance education not only means putting the study information on the net, but also means communicating between students and teachers. Communication is more importance, because of the distance. So, the question answer system is one key parts of the distance education.
The face to face can't appear in the distance education that always is seen in the traditional education. There three main question answer systems: e-mail, charting and key word searching. The shortcoming of e-mail is that this system can't give students answer in time and answer several times for the same question; Charting asks teachers on the net all the time. They all waste a lot of teacher's and answer's resource. The third of system needs students know much knowledge about the searching and increases students burden. The article imports some new technology and improves the model of question answer system to resolve those questions.
The author makes use of natural language process technology and advances a new sort of question answer system of natural language in limited domain. The system accept the natural language question and know what about the question, then the system give out a set of answers related to the question. There are three Characteristics in the question answer system, retrenching resource, intelligence, and high efficiency
Comparing with the foreign question answer system, the paper points out that the shortcoming of our system is lack of advancing technology. Firstly, the article introduces the technology about natural language process and designs the structure of the system. Secondly, the author turns up this system and discusses some key art and makes some program on the arithmetic. Finally, the paper thinks that the system is utility and is deeply researched in the future.
The main work of this paper is as the following:
Designing precept of the key of and realizing it;
Improving arithmetic of similarity;
Importing index structure and the tactic of the two level searching.
引文
1、1999年现代远程教育与技术研讨会论文集．南京
2、李爽，陈丽，国内外网上智能答疑系统比较研究．中国电化教育，2003．1
3、郑实福，刘挺，秦兵，李生，自动问答系统综述．中文信息学报．2002．6
4、张正兰，张明，多媒体技术基础及应用．河海大学出版社 2002
5、郭艳华，周昌乐，自然语言理解研究综述．杭州电子工业学院学报．2001．1
6、钱晓军，Web文本挖掘技术研究及其实现．浙江大学硕士论文，2002．3
7、罗智勇，现代汉语通用分词系统的技术与实现，北京工业大学硕士论文．2002．5
8、肖明，科技信息资源自动标引的理论与实践研究．中国科学院文献情报中心博士论文．2001．6
9、中文自然语言处理开放平台．http://www,nlp.org.cn/
10、陶跃华，王锡钢，王云爱 信息检索向量空间模型中特征提取的研究．云南师范大学学报．2003．3
11、薛翠芳，郭炳炎，汉语文本结构的自动分析．情报学报．2002．6
12、侯国锋，一个自然语言理解系统的设计和实现．计算机工程．2001．10
13、王洋，秦兵，郑实福，句子相似度计算在FAQ中的应用．http://www.nlp.org.cn/
14、董振东，董强．知网．http://www.keenage.com
15、柳泉波，黄荣怀，何克抗．智能答疑系统的设计与实现．教育技术通讯-应用开发．2000．5
16、黄河燕，李渝生，上下文相关汉语自动分词及词法预处理算法，应用科学学报．2003．9
17、申瑞民，王加俊，汤轶阳．基于Web的自动答疑系统AnserverWeb. http://www.dlc.sjtu.edu.cn
18、李水平，数据采掘技术回顾．小型微型计算机系统．1998．9
19、黄萱菁，吴立德 独立于语种的文本分类方法．2000．12
20、鲁松，白硕等：文本中词语权重计算方法的改进，2000 International Conference On MUltingual Information Processing，pp 31-36，2000
21、卜东波．聚类／分类理论研究及其在大规模文本挖掘中的应用．上海华东理工博士论
文．2000．11
22、黄德根，朱和合，王昆仑，杨元生，钟万勰，基于最长次匹配的汉语自动分词．大连理工大学学报．2003．1
23、刘亚军，徐易，高莉莎，智能答疑系统中快速定位算法的研究与实现．东南大学学报2003．6
24、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计算机工程与应用，2003．2
25、赵贇，刘亚军，智能答疑系统中语义网的研究与应用．微机发展．2003．11
26、Agirre E. and Rigau G. (1995), A proposal for word sense disambiguation using conceptual distance, in International Conference "Recent Advances in Natural Language Processing" RANLP'95, Tzigov Chark, Bulgaria
27、Li Sujian, Zhang Jian, Huang Xiong and Bai Shuo(2002), Semantic Computation in Chinese Question-Answering System, Journal of Computer Science and Technology(Accepted)
28、David D. Lewis: Feature selection and feature extraction for text categorization, In Proceedings of Speech and Natural Language Workshop, pp 212-217. Defense Advanced Research Projects Agency, Morgan Kaufmann, February 1992
29、Burke, R., Hammond, I., et al., Question Answering from Frequently-Asked Question Files: Experiences with the FAQ Finder System, Univ. of Chicago, Dept. of Computer Science Technical Report TR-97-05, 1997.
30、Robin D.Burke, Kristian J.Hammond, Vladimir A.Kulyukin.Question Answering from Frequently-Asked Question Files: Experiences with the FAQ Finder System. Technical Report 1997(5):25
31、Dagan I., Lee L. and Pereira F. (1999), Similarity-based models of word cooccurrence probabilities, Machine Learning, Special issue on Machine Learning and Natural Language, 1999
32、R. Baeza-Yates, B. Ribeiro-Neto. Modern Information Retrieval. ACM Press. Addison Wesley Longman Limited, 1999
33、Stanley B.Lippman, Josee LaJoie. C++ Primer. 潘爱民 张丽．北京：中国国力出版社，2002
34、Gudivada V N. Information retrieval on the World Wide Web. IEEE Internet Computing.1 1997,1(5):58～68
35、David D. Lewis and Marc Ringuette: A comparison of tow learning algorithms of text categorization, In Third Annual Symposium on Document Analysis and Information Retrieval, pp 81-93, Las Vegas, NV, April 11-13 1994. ISRI; Univ. of Nevada, Las Vegas
36、Liam Quin 《Open Sourse XML Database Toolkit: Resourse and Techniques for Improved Development》Whiley Computer Publishing
2、李爽，陈丽，国内外网上智能答疑系统比较研究．中国电化教育，2003．1
3、郑实福，刘挺，秦兵，李生，自动问答系统综述．中文信息学报．2002．6
4、张正兰，张明，多媒体技术基础及应用．河海大学出版社 2002
5、郭艳华，周昌乐，自然语言理解研究综述．杭州电子工业学院学报．2001．1
6、钱晓军，Web文本挖掘技术研究及其实现．浙江大学硕士论文，2002．3
7、罗智勇，现代汉语通用分词系统的技术与实现，北京工业大学硕士论文．2002．5
8、肖明，科技信息资源自动标引的理论与实践研究．中国科学院文献情报中心博士论文．2001．6
9、中文自然语言处理开放平台．http://www,nlp.org.cn/
10、陶跃华，王锡钢，王云爱 信息检索向量空间模型中特征提取的研究．云南师范大学学报．2003．3
11、薛翠芳，郭炳炎，汉语文本结构的自动分析．情报学报．2002．6
12、侯国锋，一个自然语言理解系统的设计和实现．计算机工程．2001．10
13、王洋，秦兵，郑实福，句子相似度计算在FAQ中的应用．http://www.nlp.org.cn/
14、董振东，董强．知网．http://www.keenage.com
15、柳泉波，黄荣怀，何克抗．智能答疑系统的设计与实现．教育技术通讯-应用开发．2000．5
16、黄河燕，李渝生，上下文相关汉语自动分词及词法预处理算法，应用科学学报．2003．9
17、申瑞民，王加俊，汤轶阳．基于Web的自动答疑系统AnserverWeb. http://www.dlc.sjtu.edu.cn
18、李水平，数据采掘技术回顾．小型微型计算机系统．1998．9
19、黄萱菁，吴立德 独立于语种的文本分类方法．2000．12
20、鲁松，白硕等：文本中词语权重计算方法的改进，2000 International Conference On MUltingual Information Processing，pp 31-36，2000
21、卜东波．聚类／分类理论研究及其在大规模文本挖掘中的应用．上海华东理工博士论
文．2000．11
22、黄德根，朱和合，王昆仑，杨元生，钟万勰，基于最长次匹配的汉语自动分词．大连理工大学学报．2003．1
23、刘亚军，徐易，高莉莎，智能答疑系统中快速定位算法的研究与实现．东南大学学报2003．6
24、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计算机工程与应用，2003．2
25、赵贇，刘亚军，智能答疑系统中语义网的研究与应用．微机发展．2003．11
26、Agirre E. and Rigau G. (1995), A proposal for word sense disambiguation using conceptual distance, in International Conference "Recent Advances in Natural Language Processing" RANLP'95, Tzigov Chark, Bulgaria
27、Li Sujian, Zhang Jian, Huang Xiong and Bai Shuo(2002), Semantic Computation in Chinese Question-Answering System, Journal of Computer Science and Technology(Accepted)
28、David D. Lewis: Feature selection and feature extraction for text categorization, In Proceedings of Speech and Natural Language Workshop, pp 212-217. Defense Advanced Research Projects Agency, Morgan Kaufmann, February 1992
29、Burke, R., Hammond, I., et al., Question Answering from Frequently-Asked Question Files: Experiences with the FAQ Finder System, Univ. of Chicago, Dept. of Computer Science Technical Report TR-97-05, 1997.
30、Robin D.Burke, Kristian J.Hammond, Vladimir A.Kulyukin.Question Answering from Frequently-Asked Question Files: Experiences with the FAQ Finder System. Technical Report 1997(5):25
31、Dagan I., Lee L. and Pereira F. (1999), Similarity-based models of word cooccurrence probabilities, Machine Learning, Special issue on Machine Learning and Natural Language, 1999
32、R. Baeza-Yates, B. Ribeiro-Neto. Modern Information Retrieval. ACM Press. Addison Wesley Longman Limited, 1999
33、Stanley B.Lippman, Josee LaJoie. C++ Primer. 潘爱民 张丽．北京：中国国力出版社，2002
34、Gudivada V N. Information retrieval on the World Wide Web. IEEE Internet Computing.1 1997,1(5):58～68
35、David D. Lewis and Marc Ringuette: A comparison of tow learning algorithms of text categorization, In Third Annual Symposium on Document Analysis and Information Retrieval, pp 81-93, Las Vegas, NV, April 11-13 1994. ISRI; Univ. of Nevada, Las Vegas
36、Liam Quin 《Open Sourse XML Database Toolkit: Resourse and Techniques for Improved Development》Whiley Computer Publishing