不完整数据分类知识发现算法研究
|
摘要
分类知识发现是数据挖掘的基本任务,也是知识发现中最重要的目标之一。据统计,在机器学习和数据挖掘应用过程中不完整数据的理解需要花费大量的时间和精力,因此不完整数据处理是现实世界中分类知识挖掘必须认真对待的重要问题。本文以提高不完整数据的分类知识发现算法性能为切入点,探索充分利用不完整数据集中隐含信息和提高数据挖掘效率的途径。本文具体的研究工作如下：
(1)出于提高算法分类正确率的目的,针对朴素信念分类算法忽略属性变量的投票权重,提出了基于相关系数的加权保守推理规则。
此规则尝试用权重量化不完整数据中属性变量与类别之间的相关程度,基于此思路改进了朴素信念分类算法,并在国际公开的数据集上与现有的主要分类算法进行了分类对比实验。实验结果表明在不需要对不完整数据进行填充处理,并由此避免因不合理填充方法引起数据倾斜的情况下,该算法能够充分学习不完整数据中蕴含的隐藏信息,学习性能优于朴素信念分类和朴素贝叶斯分类算法,在某些数据集上与支持向量机不相上下。尤其是在朴素贝叶斯分类准确率表现不佳的样本上,不完整数据条件下的加权朴素信念分类算法得到了较好的分类结果。
(2)针对目前半监督分类算法中未考虑缺失属性数据项隐含信息和算法复杂度高的情况,本文提出两阶段半监督加权朴素信念分类模型。
此模型将半监督分类过程分为两个阶段的加权朴素信念分类,与直推支持向量机和在国际公开标准数据集上的对比实验表明两阶段半监督加权朴素信念分类模型有效地减少了分类时间,而在其能够明确分类样本上的正确率与直推支持向量机相当。
(3)为了增强朴素信念分类算法的鲁棒性,提高其明确分类样本比例低的情况,本文提出基于放松区间优势的不完整数据分类模型。
此模型在放松区间优势定义的基础上改进了朴素信念分类,在国际公开标准数据集上的对比实验表明此模型在大多数的数据集上起到了改善朴素信念分类和加权朴素信念分类算法明确分类样本比例的作用,有利于做出确切的分类判断,同时保证了较高的分类正确率,总体分类性能优于朴素信念分类、加权朴素信念分类、朴素贝叶斯算法和最近邻法,但是否优于支持向量机要观察其在不同数据集上的表现。
最后,本文将加权朴素信念分类、两阶段加权朴素信念半监督分类算法和放松区间优势朴素信念分类算法分别应用于文体风格识别不完整数据集,取得了较理想的实验结果,验证了算法的有效性。
Category knowledge discovery is the fundamental task of data mining and one of the most important goals in knowledge discovery. According to statistics, the understanding of incomplete data in machine learning and data mining application process need to spend a lot of time and effort. So the processing of incomplete data from real world should to be taken seriously an important issue in classification knowledge discovery. As the starting point to explore the classification of incomplete data, this paper focus on the full use of hidden information in incomplete data sets and efficient way to improve data mining.The detailed contents of the research are as follows:
(1) The weighted conservative inference rule based on correlation coefficient is proposed. This rule tries to make use of the correlation coefficient to quantitative analysis the relationship between the attributes and the categories. Based on this idea, the weighted Naive Credal classifier is proposed and tested on the international public data sets. Compared with Naive Bayes classifier and Naive Credal classifier, this algorithm has better learning performance. On some datasets, the weighted Naive Credal classifier is comparable with the support vector machine. Compared with other existing classification algorithms, the weighted Naive Credal classifier performs better owning to the full use of the hidden information in incomplete data.
(2) This paper presents a two-stage semi-supervised weighted Naive Credal classification model. For the ignoring of the implicit information in incomplete data and the high complexity of current semi-supervised classifiers, in this model the semi-supervised classification process is divided into two weighted Naive Credal classification stages. Compared with transductive Support Vector Machine (TSVM), this algorithm has lower time complexity and almost the same accuracy.
(3) This paper presents a Naive Credal classifier based on relaxed conservative inference rule for incomplete data. For the low proportion of determinate classified samples, the definition of interval advantages is relaxed in this model. Compared with Naive Credal classifier and weighted Naive Credal classifier, this algorithm effectively increases the proportion of determinate classified samples and almost the same accuracy. Overall this algorithm has better classification performance than Naive Bayes classifier, nearest neighbor method, Naive Credal classifier and weighted Naive Credal classifier. But if this algorithm has better performance than support vector machine depends on their performance on different data sets.
Finally, the weighted Naive Credal classifier, the two-stage semi-supervised weighted Naive Credal classifier and the relaxed Conservative Inference Rule based weighted Naive Credal classifier are applied to style identification dataset. The validity of the algorithm is verified by the better experimental results compared with the main of existing classification algorithms.
(1)出于提高算法分类正确率的目的,针对朴素信念分类算法忽略属性变量的投票权重,提出了基于相关系数的加权保守推理规则。
此规则尝试用权重量化不完整数据中属性变量与类别之间的相关程度,基于此思路改进了朴素信念分类算法,并在国际公开的数据集上与现有的主要分类算法进行了分类对比实验。实验结果表明在不需要对不完整数据进行填充处理,并由此避免因不合理填充方法引起数据倾斜的情况下,该算法能够充分学习不完整数据中蕴含的隐藏信息,学习性能优于朴素信念分类和朴素贝叶斯分类算法,在某些数据集上与支持向量机不相上下。尤其是在朴素贝叶斯分类准确率表现不佳的样本上,不完整数据条件下的加权朴素信念分类算法得到了较好的分类结果。
(2)针对目前半监督分类算法中未考虑缺失属性数据项隐含信息和算法复杂度高的情况,本文提出两阶段半监督加权朴素信念分类模型。
此模型将半监督分类过程分为两个阶段的加权朴素信念分类,与直推支持向量机和在国际公开标准数据集上的对比实验表明两阶段半监督加权朴素信念分类模型有效地减少了分类时间,而在其能够明确分类样本上的正确率与直推支持向量机相当。
(3)为了增强朴素信念分类算法的鲁棒性,提高其明确分类样本比例低的情况,本文提出基于放松区间优势的不完整数据分类模型。
此模型在放松区间优势定义的基础上改进了朴素信念分类,在国际公开标准数据集上的对比实验表明此模型在大多数的数据集上起到了改善朴素信念分类和加权朴素信念分类算法明确分类样本比例的作用,有利于做出确切的分类判断,同时保证了较高的分类正确率,总体分类性能优于朴素信念分类、加权朴素信念分类、朴素贝叶斯算法和最近邻法,但是否优于支持向量机要观察其在不同数据集上的表现。
最后,本文将加权朴素信念分类、两阶段加权朴素信念半监督分类算法和放松区间优势朴素信念分类算法分别应用于文体风格识别不完整数据集,取得了较理想的实验结果,验证了算法的有效性。
Category knowledge discovery is the fundamental task of data mining and one of the most important goals in knowledge discovery. According to statistics, the understanding of incomplete data in machine learning and data mining application process need to spend a lot of time and effort. So the processing of incomplete data from real world should to be taken seriously an important issue in classification knowledge discovery. As the starting point to explore the classification of incomplete data, this paper focus on the full use of hidden information in incomplete data sets and efficient way to improve data mining.The detailed contents of the research are as follows:
(1) The weighted conservative inference rule based on correlation coefficient is proposed. This rule tries to make use of the correlation coefficient to quantitative analysis the relationship between the attributes and the categories. Based on this idea, the weighted Naive Credal classifier is proposed and tested on the international public data sets. Compared with Naive Bayes classifier and Naive Credal classifier, this algorithm has better learning performance. On some datasets, the weighted Naive Credal classifier is comparable with the support vector machine. Compared with other existing classification algorithms, the weighted Naive Credal classifier performs better owning to the full use of the hidden information in incomplete data.
(2) This paper presents a two-stage semi-supervised weighted Naive Credal classification model. For the ignoring of the implicit information in incomplete data and the high complexity of current semi-supervised classifiers, in this model the semi-supervised classification process is divided into two weighted Naive Credal classification stages. Compared with transductive Support Vector Machine (TSVM), this algorithm has lower time complexity and almost the same accuracy.
(3) This paper presents a Naive Credal classifier based on relaxed conservative inference rule for incomplete data. For the low proportion of determinate classified samples, the definition of interval advantages is relaxed in this model. Compared with Naive Credal classifier and weighted Naive Credal classifier, this algorithm effectively increases the proportion of determinate classified samples and almost the same accuracy. Overall this algorithm has better classification performance than Naive Bayes classifier, nearest neighbor method, Naive Credal classifier and weighted Naive Credal classifier. But if this algorithm has better performance than support vector machine depends on their performance on different data sets.
Finally, the weighted Naive Credal classifier, the two-stage semi-supervised weighted Naive Credal classifier and the relaxed Conservative Inference Rule based weighted Naive Credal classifier are applied to style identification dataset. The validity of the algorithm is verified by the better experimental results compared with the main of existing classification algorithms.
引文
[I]U. Fayyad, Gregory Piatetsky-Shapiro, Padhraic Smyth. From data mining to knowledge discovery:an overview. Advances in knowledge discovery and data mining. American Association for Artificial Intelligence.1996:1-34.
[2]U. Fayyad, G. Piatetsky Shapiro, Padhraie Smyth. Knowledge Discovery and Data Mining:Towards a Unifying Framework. Proceedings of Second International Conference on Knowledge Discovery and DataMining (KDD 96), AAAIPress,1996.
[3]U. M. Feyyad. Data mining and knowledge discovery:making sense out of data. IEEE Expert.Volume:11. Issue:5.2002:20-25.
[4]Christian Bohm and Florian Krebs.The k-Nearest Neighbour Join:Turbo Charging the KDD Process. Knowledge and Information Systems. Volume 6, Number 6.2004:728-749.
[5]Antonio Congiusta, Domenico Talia and Paolo Trunf io. Parallel and Grid-Based Data Mining-Algorithms, Models and Systems for High-Performance KDD. Data Mining and Knowledge Discovery Handbook 2010:1009-1028.
[6]Zubcoff, J. and Trujillo, J. Conceptual Modeling for Classification Mining in Data Warehouses. DaWaK. Krakow. Poland,2006:566-575.
[7]Derya Birant, Alp Kut. ST-DBSCAN:An algorithm for clustering spatial-temporal data. Data & Knowledge Engineering. Volume 60, Issue 1, January 2007:208-221.
[8]Jain A.K., Dubes, R. C. Algorithms for Clustering Data. Prentice-Hall, Inc. USA.1988.
[9]Jain, A. K., Murty M.N., Flynn P.J. Data Clustering:A Review, ACM Computing Surveys, Vol 31, No.3,1999:264-323.
[10]Jose Zubcoff, uan Trujillo. An approach for the conceptual modeling of clustering mining in the KDD process IADIS European Conference. Data Ming 2007.
[11]Abonyi Janos, Feil Balazs. Cluster Analysis for Data Mining and System Identification. Dordrecht:Springer,2007:315-317.
[12]Christoph Schmitz, Andreas Hotho, Robert Jaschke and Gerd Stumme. Mining Association Rules in Folksonomies.Data Science and Classification. Studies in Classification, Data Analysis, and Knowledge Organization, Part Ⅵ,2006:261-270.
[13]Abdullah Uz Tansel, Susan P. Imberman. Discovery of Association Rules in Temporal Databases, Information Technology:New Generations. Third International Conference on International Conference on Information Technology (ITNG'07),2007. pp.371-376.
[14]Philippe Lenca, Patrick Meyer, Benoit Vaillant, Stephane Lallich. On selecting interestingness measures for association rules:User oriented description and multiple criteria decision aid.Computing, Artificial Intelligence and Information Management. European Journal of Operational ResearchVolume 184, Issue 2,16 January 2008, Pages 610-626.
[15]Chien-Liang Wu, Jia-Ling Koh, Pao-Ying An. Improved Sequential Pattern Mining Using an Extended Bitmap Representation. Database and Expert Systems Applications. Lecture Notes in Computer Science,2005, Volume 3588.2005:776-785.
[16]Zhenglu Yang, Kitsuregawa, M. LAPIN-SPAM:An Improved Algorithm for Mining Sequential Pattern. Data Engineering Workshops,2005:1221-1222.
[17]Xilu Wang, Weill Yao. Sequential Pattern Mining:Optimum Maximum Sequential Patterns and Consistent Sequential Patterns. Integration Technology. ICIT'07,2007:365-368.
[18]Lior Rokach, Oded Maimon.Data Mining with Decision Trees:Theroy and Applications. Machine Perception and Artificial Intelligence archive. World Scientific.2008:239-244.
[19]San Jose, California. Mining optimal decision trees from itemset lattices. International Conference on Knowledge Discovery and Data Mining archive. Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining,2007:530-539.
[20]朱明.数据挖掘.合肥中国科技大学出版社,2002.
[21]Jiawei Han, Micheline Kamber:Data Mining Concepts and Techniques, Morgan Kaufmann Publishers,2000.
[22]Lakshminarayan K. et al. Imputation of missing data in industrial databases, Applied Intelligence.1999,11:259-275.
[23]Dixon J. K. Pattern recognition with partly missing data. IEEE Transactions on Systems, Man and Cybernetics, Vol.9, No.10,1979:617-621.
[24]Zhang C., Yang Q., Liu B., Intelligent Data Preparation. IEEE Transactions on Knowledge and Data Engineering.17(9) 2005:1163-1165.
[25]Marvin L. Brown, John F. Kros, Data mining and the impact of missing data. Industrial Management & Data Systems.103(8),2003:611-621.
[26]Little R., Rubin D. Statistical Analysis with Missing Data. (Second Edition). John Wiley and Sons, New York.2002.
[27]Abraham, Todd W, Russell. Missing data:a review of current methods and applications in epidemiological research.Current Opinion in Psychiatry.2004(04), vol.17 no.4,pp.315-321
[28]M. Ramoni, P.Sebastiani. Robust Bayes classifiers. Artificial Intelligence.2001, 125(1-2):209-226.
[29]Zaffalon M. The naive credal classifier.Journal of Statistical Planning and Inference Volume 105, Issue 1,15 June 2002:5-21.
[30]陈景年.选择性贝叶斯分类算法研究.博士学位论文.北京：北京交通大学2008.
[31]Hsu C, Huang H, Wong T. Implications of the Dirichlet Assumption for Discretization of Continuous Variables in Naive Bayesian Classifiers. Machine Learning.2003:53(3),235-263.
[32]Manski CF. Partial Identification of Probability Distributions. Springer-Verlag, New York.2003.
[33]Brown, M. L. and Kros, J. F. Industrial Management and Data Systems,103(8).2003: 611-621.
[34]Huang,X. L. A pseudo-nearest-neighbor approach for missing data recovery o Gaussian random data sets. Pattern Recognition Letters,2002(23):1613-1622.
[35]Grzymala-Busse, J. W., and Fu, M. A comparison of several approaches to missing attribute values in data mining. In:Proc of the 2nd Int'Conf on Rough Sets and Current.2000.
[36]Jonsson P., Claes W. An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[37]Yu, J. General C-means clustering model. IEEE Tranctions on Pattern Analysis and Machine Intelligence, vol.271, no.8,2005:1197-1211.
[38]Zhang, S. C., Qin,Z.X., Sheng, S. L. and Ling, C. L. "missing is useful":missing valucs incostsensitive decision trees. IEEE Transactions on Knowledge and Data Engineering, Vol.17 No.12.2005.
[39]Pearson R. K. The Problem of Disguised Missing Data ACM SIGKDD Explorations Newsletter archive Volume 8, Issue 1, June.2006.
[40]Rubin, D. B.Multiple Imputation for Nonresponse in Surveys, Wiley:New York. 1987.
[41]SL. Lauritzen. The EM Algorithm for Graphical Association models with Mssing Data Computational Statistics and Data Analysis,1995:191-201.
[42]Dempster, A. P., Laird, N. M. Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Vol.39,1977:1-38.
[43]N. Friedman, D, Geiger&M. Goldszmidt. Bayesian network classifiers. Machine Learning,1997,29(2-3):131-163.
[44]Geman S., Geman D. Stochastic relaxation, Gibbs distributionand the Bayesia restoration of images. IEEE Trans onPattern Analysis and Machine Intelligence,1984, 6:721-742.
[45]Meng, X.L., Rubin, D. B. Maximum likelihood estimation via the ECM algorithm A general framework. Biometrika,1993,80:267.
[46]Sebastian, P., Ramoni, M. Bayesian inference with missing data using bound an collapse. Journal of Computational and Graphical Statistics,2000,9(4):779.
[47]Joseph G. Ibrahim, Hongtu Zhu, Niansheng Tang. Model Selection Criteria for Missing-Data Problems Using the EM Algorithm.J Am Stat Assoc.2008 December 1; 103 (484):1648-1658.
[48]Schafer J., Graham J. Missing data:Our view of the state of the art. Psychological Methods,2002:147-177.
[49]Roth P. Missing data:A conceptual review for applied psychologists. Personnel Psychology,47.1994:537-560.
[50]Sarle W. S. "Prediction with Missing Inputs, "in Wang, P. P. (ed.), JCIS'98 Proceedings, Vol II, Research Triangle Park, NC,1998:399-402.
[51]Allison P. Missing Data. Sage Publication, Inc,2001.
[52]Wang S. C., Yuan S. M. Research on learing bayesian networks structure with missing data, Journal of Software, vol.15 2004:1024-1048.
[53]Wei W., Ying T. A generic Neural Network Approach for Filling Missing Data in Data Ming, Proceedings of the IEEE International Conference on Systems, Man and Cybernetics,2003.vol.1:862-867.
[54]Jonsson P., Claes W. An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[55]VapnikV N.Statistical Learning Theory. Wiley, New York,1998.
[56]Wang, Q. D et al. A rough set based clustering algorithm and the information theoretical approach to refine clusters. ICMM 2004.
[57]雷蕾,吴乃君,刘鹏,刘兰娟灵敏度分析：分类器中的缺失数据.管理学报.2005(9)：153-156.
[58]Cios KJ, Kurgan L. Trends in Data Mining and Knowledge Discovery. Advanced Techniques in Knowledge Discovery and Data Mining. Springer London.2005:1-26
[59]Marvin L Brown, John F Kros. Data Mining and The Impact of Missing Data Data Mining:Opportunities and Challenges, Idea Group Publishing, USA,2003:611-621.
[60]Quinlan, J. R. Unknown attributes values. In Programs for Machine Learning. Morgan Kaufmann,1993:27-32.
[61]Wang, S. C., and Yuan, S. M. Research on learing bayesian networks structure with missing data, Journal of Software, vol.15.2004:1024-1048.
[62]Jonsson P., Claes W.An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[63]Shahshahani B, Landgrebe D. The effect of unlabeled samples in reducing the small sample size problem and mitigating the Hughes phenomenon. IEEE Transactions of Geoscience and Remote Sensing,1994,32(5):1087-1095.
[64]0 Chapelle, B Scholkopf, A Zien. Semi-supervised learning. Cambridge MA: MITPress,2006.
[65]Avrim Blum, Tom Mitchell. Combining labeled and unlabeled data with co-training. Proceedings of the eleventh annual conference on Computational learning theory, NewYork:ACMPress,1998:92-100.
[66]Yarowsky D.Unsupervised word sense disambiguation rivaling supervised methods. Proceedings of the 33rd Annual Meeting of the Association for Computatonal Linguistics,1995:189-196.
[67]Riloff E, Wiebe J, Wilson T. Learning subjective nouns using extraction pattern boot strapping. Proceedings of the Seventh Conference on Natural Language Learning (CoNLL-2003),2003.
[68]Zhou Z H, Zhan D C, Yang Q. Semi-supervised learning with very few labeled training examples. Association for the Advancement of Artificial Intelligence.http: //www. aaai. org.
[69]Nigam K, McCallum, A K, Thrun s, et al.Text classification from labeled and unlabeled documents using EM. Machine Learning,2000(39):103-134.
[70]Nigam K. Using unlabeled data to improve text classification (Technical Report CMU-CS-01-126). Carnegie Mellon University, Doctoral Dissertation,2001.
[71]Baluja S. Probabilistic modeling for face orientation discrimination:leaming from labeled and unlabeled data. Neural Information Proeessing Systems,1998.
[72]Fujino A, Ueda N, Saito K.A hybrid generative/discriminative approach to semi-supervised classifier design. The Twentieth National Conference on Artificial Intelligence(AAAI05),2005.
[73]Joachims T. Transductive inference for text classification using support vector maehines. Proeeedings of the 16th International Conference on Machine Learning, San Fransiseo,1999:200-209.
[74]Nigam K, Ghani R. Analyzing the effectiveness and applicability of co-training. Proceedings of the 9th ACM International Conference on Information and Knowledge Management (CIKM2000),Mclean, VA,2000:86-93.
[75]Goldman S, Zhou Y. Enhancing supervised learning with unlabeled data. Proceedings of the 17th International Conference on Machine Learning (ICML2000). San Francisco:Morgan Kaufmann Publishers,2000:327-334.
[76]Zhou Z H, Li M. Tri-training:Exploiting unlabeled data using threclassifiers. IEEE Transactions on Knowledge and Data Engineering,2005,17 (11):1529-1541.
[77]J. Zhu.Semi-Supervised Learning Literature Survey.Computer Sciences, University of Wisconsin-Madison, Technical report 1530,2008.
[78]X. Zhu.Semi-supervised learning with graphs. Doctoral dissertation (CMU-LTI-05-192), Carnegie Mellon University.2005.
[79]Z. Y. Niu, D. H. Ji, C. L. Tan. Word sense disambiguation using label propagation based semi-supervised learning. In Proceedings of the ACL05.2005:395-402.
[80]A.B.Goldberg, X.Zhu. Seeing stars when there aren't many stars:Graph-based semi-supervised learning for sentiment categorization. In HLTNAACL 2006 Workshop on Textgraphs:Graph-based Algorithms for Natural Language Processing. NY, USA.2006. LNCS5012:1006-1014.
[81]L.Grady, G. Funka-Lea. Multi-label image segmentation for medical applications based on graph-theoretic electrical potentials. In ECCV 2004 workshop.17. Cambridge, MA:MIT Press.2004:230-245.
[82]BelkinM, Matveeval, NiyogiP. Regression and regularization on large graphs. Proeeedings of the 17th Annual Conference on Learning Theory, NewYork:ACM Press, 2004:185-192.
[83]P. Domingos, M. Pazzani. On the Optimality of the Simple Bayesian Classifier under Zero-One Loss. Machine Leaming,1997,29(2-3):103-130.
[84]M.J. Pazzani. Searching for Dependencies in Bayesian Classifiers. Learning from Data, Artificial Intelligence and Statiscics. New York, Springer Verlag,1996:239-248.
[85]Zi jian Zheng. Naive bayesian classifier committees. In Proceedings of ECML'98, Berlin:Springer Verlag,1998:196-207.
[86]Zhu XJ. Semi-supervised learning with graphs. USA:Carnegie Mellon University. 2006:1-89.
[87]陈锦秀,姬东鸿.基于图的半监督关系抽取.软件学报.2008,19(11)：2843-2852.
[88]X Zhu, Z Ghahramani, J Lafferty. Semi-supervised learning using Gaussian fields and harmonic functions. Proeeedings of the 20th International conference on Machine Learning (ICML2003), Washington DC,2003:912-919.
[89]D Zhou,O Bousquet, TN Lal, J Weston. Learning, with local and global consisteney. Advances in Neural Information Proeessing Systems16, Cambridge, MA:MIT Press,2004:321-328.
[90]M Belkin, P Niyogi, V Sindwani. On manifold regularization. Proeeedings of the 10th International Workshop on Artificial Intelligence and Statistics(AISTATS2005), 2005:17-24.
[91]Chapelle O, Chi M, Zien A. A continuation method for semi-supervised SVMs.23rd International Conference on Machine Learning. USA, Pittsburgh,2006:1148-1157.
[92]Chapelle O, Scholkopf B, Zien A. Semi-Supervised Learning. Massachusetts.MIT Press,2006:33-543.
[93]Chapelle O,Zien A.Semi-supervised classification by low density separation Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics. Barbados.2005:57-64.
[94]Sindhwani V,Keerthi S, Chapelle 0. Deterministic annealing for semi-supervised kernelmachines. Proceedings of the 23rd International Conference on Machine Learning. USA:Association for Computing Machinery.2006:841-848.
[95]Chapelle 0, Sindhwani V, Keerthi SS. Branch and bound for semi-supervised support vectormachines. Advances in Neural Information Processing Systems, USA:MIT Press.2006:217-224.
[96]陈毅松.基于支持向量机的渐进直推式分类学习算法.软件学报,2003,3：459-460.
[97]Reshmi Malhotra, DK Malhotra. Differentiating between good credits and bad credits using neuro-fuzzy systems. Computing artificial intelligence and information technology.2002,136(1):190-211.
[98]Murphy P M, Aha D W. UCI Machine Learning Repository,2009-1-9. http://www. ics. uci. edu/～mlearn.
[99]WEKA. http://www. cs. waikato.ac.nz/ml/weka/.2009-01-11.
[100]Giorgio Corani, Marco Zaffalon. JNCC2:The Java Implementation Of Naive Credal Classifier 2. Journal of Machine Learning Research.2008,9:2695-2698.
[101]Lann. Stylometry and method. The case of Euripides. Literary and linguistic computing,1995, (10):271-278.
[102]Baayen, Harald, Hans, van Halteren & Fiona Tweedie. Outside the Cave of Shadows: Using Syntactic Annotation to Enhance Authorship Attribution. Literary and Linguistic Computing,1996,11(3):121-131.
[103]Stamatos, Efstathios, Nikos Fakotakis, George Kokkinakis. Automatic Text Categorization in Terms of Genre and Author. Computational Linguistics,2000,26(4): 471-495.
[104]Michael Gamon. Linguistic correlates of style:authorship classification with deep linguistic analysis features. In Proceeding 20th International Conference of Computational Linguistics, Geneva,2004:611-617.
[105]霍跃红,典籍英译译者文体分析与译者文本识别.博士学位论文.大连：大连理工大学,2010.
[106]金奕江,孙晓明,马少平.因特网上的写作风格鉴别.广西师范大学学报(自然科学版).2003,21(1)：62-66.
[107]马建斌.基于SVM的中文电子邮件作者身份挖掘技术研究.硕十学位论文.保定：河北农业大学,2004.
[108]常淑惠.基于写作风格的中文邮件作者身份识别技术研究.硕士学位论文.保定：河北农业大学,2005.
[109]武晓春,黄萱菁,吴立德.基于语义分析的作者身份识别方法研究.中文信息学报.2006,20(6)：61-68.
[110]Mat sura, Kanada. Authorship Detection of Sentences by 8 Japanese Modern Authors via N-gram Distribution. IPSJ SIG Notes,2000-NL-137:1-8.
[111]Yoshida, Nobesawa, Saito. Effective Features of Authorship Identification. IPSJ SIG Notes,2001-NL-145:83-90.
[112]J. Hoorn, S. Frank, W. Kowalczyk, F. van der Ham. Neural netword indentif ication of poets using letter sequences. Literary and Linguistic Computing,14(3):311-338.
[113]Cagatay Catal, Kemalettin Erbakici, Yasar Erenler. Computer-Based authorship attribution for Turkish documents. IJCI proceeding of Intl. Ⅻ. Tukish Symoposium on Artificail Interlligence and Neural Networks.2003.
[114]Zhao Ying, Zobel Justin. Effective and Scalable Authorship Attribution Using Function Words. Lecture Notes in Computer Science.2005:174-189.
[115]易勇.计算机辅助诗词创作中的风格辨析及联语应对研究.博士学位论文.重庆：重庆大学,2005.
[116]张全,张运良,袁毅.利用语言概念表示的作者写作风格分类研究.第七届中文信息处理国际会议论文集.2007：460-464.
[117]张运良,朱礼军,乔晓东,张全.基于句类特征的作者写作风格分类研究.计算机工程与应用.2009.22
[118]吴春龙周昌乐,基于频繁关键字共现的诗词风格分类模型研究,厦门大学学报(自然科学版).2008(1)：41-44.
[119]Miller D J, Uyar H S. A mixture of experts classifier with learning based on both labelled and unlabelled data. Mozer M, Jordan M I, Petsche T, Advances in Neural Information Processing Systems 9, Cambridge, MA:MIT Press,1997:571-577.
[120]T.Zhang, F. J. Oles.A probability analysis on the value of unlabeled data for classification problems. In:Proceedings of the 17th International Conference on Machine Learning (ICML'00), San Francisco, CA,2000:1191-1198.
[121]0. Chapelle and B. Schokopf and A. Zien. Semi-Supervised Learning. MIT Press, Cambridge, MA,2006.
[122]周志华,王珏.机器学习及其应用.北京：清华大学出版社,2007.
[123]Xiaojin Zhu. Semi-supervised learning literature survey. TR-1530. University of Wisconsin-Madison Department of Computer Science.2005.
[124]Xiaojin Zhu. Semi-Supervised Learning Tutorial. Tutorial of ICML07,2007.
[125]W. Wang, Z.-H. Zhou. Analyzing co-training style algorithms. In:Proceedings of the 18th European Conference on Machine Learning (ECML'07), Warsaw, Poland,2007.
[126]Zhong S. Semi-Supervised model-based document clustering:A comparative study. Machine Learning,2006,65(1):3-29.
[127]Bouchachia A, Pedrycz W. Data clustering with partial supervision. Data Mining and Knowledge Discovery,2006,12(1):47-78.
[128]邓超,郭茂祖.基于自适应数据剪辑策略的Tri-training算法.计算机学报.2007(9)：1213-1226.
[129]Guifa Teng, Yihong Liu, Jianbin Ma, Fang Wang, Huiting Yao. Improved Algorithm for Text Classification Based on TSVM. Proceedings of the First International Conference on Innovative Computing, Information and Control.2006:55-58.
[130]YE WANG, Shang-Teng HUANG.Training TSVM with the proper number of positive samples. Pattern recognition letters. Elsevier,2005:1414,2187-2194.
[131]廖东平,姜斌,魏玺章,黎湘,庄钊文.一种快速的渐进直推式支持向量机分类学习算法.系统工程与电子技术,2007,29(1)：87-91.
[132]Ramesh Agarwal, Mahesh V. Joshi. IBM ResearchReport. PNrule:A new Framework for Learning Classifier Models in Data Mining (A Case-Study in Network Intrusion Detection). In Proceedings of First SIAM Conference on DataMining, Chieago, April,2001.
[133]Goldman S, Zhou Y. Enhancing supervised learning with unlabeled data. In Proceedings of 17th International Conference on Machine Learning (ICMLOO),2000, 327-334.
[134]Zhou Y, Goldman S. Democratic co-learing.In Proceedings of thel6th IEEE International Conference on Tools with Artificial Intelligence,2004.
[135]郝秀兰,胡运发.半监督的文本分类——两阶段协同学习.小型微型计算机系统.Vo1.30 No.10,2009：1921-1924.
[136]Giorgio Corani, Marco Zaffalon. JNCC2:The Java Implementation of Naive Credal Classifier 2. Journal of Machine Learning Research.2008,9:2695-2698.
[137]Joachims. Multi-class Support Vector Machine. http://www.cs.cornell.edu/ People/tj/svm_light/svm_multiclass. html.2008.8.
[138]Mccallum. A Mccallum Rainbow. http://www. cs. cmu. edu/~mccallum/bow/ rainbow.2008.9.
[139]WALLEY P. Statistical Reasoning with Imprecise Probabilities. London:Chapman and Hall.1991.
[140]IGOR 0 K, YEVGENY V F. Imprecise system reliability. International Journal of Systems Science,2001,32 (4):487-493.
[141]Weichselberger K. The theory of interval-probability as a unifying concept for uncertainty.lnernational Journal of Approximate Reasonging,2000, Vol.24:149-170.
[172]冯向前.区间数不确定多属性决策方法研究.博士学位论文.南京：南京航空航天大学,2007.
[2]U. Fayyad, G. Piatetsky Shapiro, Padhraie Smyth. Knowledge Discovery and Data Mining:Towards a Unifying Framework. Proceedings of Second International Conference on Knowledge Discovery and DataMining (KDD 96), AAAIPress,1996.
[3]U. M. Feyyad. Data mining and knowledge discovery:making sense out of data. IEEE Expert.Volume:11. Issue:5.2002:20-25.
[4]Christian Bohm and Florian Krebs.The k-Nearest Neighbour Join:Turbo Charging the KDD Process. Knowledge and Information Systems. Volume 6, Number 6.2004:728-749.
[5]Antonio Congiusta, Domenico Talia and Paolo Trunf io. Parallel and Grid-Based Data Mining-Algorithms, Models and Systems for High-Performance KDD. Data Mining and Knowledge Discovery Handbook 2010:1009-1028.
[6]Zubcoff, J. and Trujillo, J. Conceptual Modeling for Classification Mining in Data Warehouses. DaWaK. Krakow. Poland,2006:566-575.
[7]Derya Birant, Alp Kut. ST-DBSCAN:An algorithm for clustering spatial-temporal data. Data & Knowledge Engineering. Volume 60, Issue 1, January 2007:208-221.
[8]Jain A.K., Dubes, R. C. Algorithms for Clustering Data. Prentice-Hall, Inc. USA.1988.
[9]Jain, A. K., Murty M.N., Flynn P.J. Data Clustering:A Review, ACM Computing Surveys, Vol 31, No.3,1999:264-323.
[10]Jose Zubcoff, uan Trujillo. An approach for the conceptual modeling of clustering mining in the KDD process IADIS European Conference. Data Ming 2007.
[11]Abonyi Janos, Feil Balazs. Cluster Analysis for Data Mining and System Identification. Dordrecht:Springer,2007:315-317.
[12]Christoph Schmitz, Andreas Hotho, Robert Jaschke and Gerd Stumme. Mining Association Rules in Folksonomies.Data Science and Classification. Studies in Classification, Data Analysis, and Knowledge Organization, Part Ⅵ,2006:261-270.
[13]Abdullah Uz Tansel, Susan P. Imberman. Discovery of Association Rules in Temporal Databases, Information Technology:New Generations. Third International Conference on International Conference on Information Technology (ITNG'07),2007. pp.371-376.
[14]Philippe Lenca, Patrick Meyer, Benoit Vaillant, Stephane Lallich. On selecting interestingness measures for association rules:User oriented description and multiple criteria decision aid.Computing, Artificial Intelligence and Information Management. European Journal of Operational ResearchVolume 184, Issue 2,16 January 2008, Pages 610-626.
[15]Chien-Liang Wu, Jia-Ling Koh, Pao-Ying An. Improved Sequential Pattern Mining Using an Extended Bitmap Representation. Database and Expert Systems Applications. Lecture Notes in Computer Science,2005, Volume 3588.2005:776-785.
[16]Zhenglu Yang, Kitsuregawa, M. LAPIN-SPAM:An Improved Algorithm for Mining Sequential Pattern. Data Engineering Workshops,2005:1221-1222.
[17]Xilu Wang, Weill Yao. Sequential Pattern Mining:Optimum Maximum Sequential Patterns and Consistent Sequential Patterns. Integration Technology. ICIT'07,2007:365-368.
[18]Lior Rokach, Oded Maimon.Data Mining with Decision Trees:Theroy and Applications. Machine Perception and Artificial Intelligence archive. World Scientific.2008:239-244.
[19]San Jose, California. Mining optimal decision trees from itemset lattices. International Conference on Knowledge Discovery and Data Mining archive. Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining,2007:530-539.
[20]朱明.数据挖掘.合肥中国科技大学出版社,2002.
[21]Jiawei Han, Micheline Kamber:Data Mining Concepts and Techniques, Morgan Kaufmann Publishers,2000.
[22]Lakshminarayan K. et al. Imputation of missing data in industrial databases, Applied Intelligence.1999,11:259-275.
[23]Dixon J. K. Pattern recognition with partly missing data. IEEE Transactions on Systems, Man and Cybernetics, Vol.9, No.10,1979:617-621.
[24]Zhang C., Yang Q., Liu B., Intelligent Data Preparation. IEEE Transactions on Knowledge and Data Engineering.17(9) 2005:1163-1165.
[25]Marvin L. Brown, John F. Kros, Data mining and the impact of missing data. Industrial Management & Data Systems.103(8),2003:611-621.
[26]Little R., Rubin D. Statistical Analysis with Missing Data. (Second Edition). John Wiley and Sons, New York.2002.
[27]Abraham, Todd W, Russell. Missing data:a review of current methods and applications in epidemiological research.Current Opinion in Psychiatry.2004(04), vol.17 no.4,pp.315-321
[28]M. Ramoni, P.Sebastiani. Robust Bayes classifiers. Artificial Intelligence.2001, 125(1-2):209-226.
[29]Zaffalon M. The naive credal classifier.Journal of Statistical Planning and Inference Volume 105, Issue 1,15 June 2002:5-21.
[30]陈景年.选择性贝叶斯分类算法研究.博士学位论文.北京：北京交通大学2008.
[31]Hsu C, Huang H, Wong T. Implications of the Dirichlet Assumption for Discretization of Continuous Variables in Naive Bayesian Classifiers. Machine Learning.2003:53(3),235-263.
[32]Manski CF. Partial Identification of Probability Distributions. Springer-Verlag, New York.2003.
[33]Brown, M. L. and Kros, J. F. Industrial Management and Data Systems,103(8).2003: 611-621.
[34]Huang,X. L. A pseudo-nearest-neighbor approach for missing data recovery o Gaussian random data sets. Pattern Recognition Letters,2002(23):1613-1622.
[35]Grzymala-Busse, J. W., and Fu, M. A comparison of several approaches to missing attribute values in data mining. In:Proc of the 2nd Int'Conf on Rough Sets and Current.2000.
[36]Jonsson P., Claes W. An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[37]Yu, J. General C-means clustering model. IEEE Tranctions on Pattern Analysis and Machine Intelligence, vol.271, no.8,2005:1197-1211.
[38]Zhang, S. C., Qin,Z.X., Sheng, S. L. and Ling, C. L. "missing is useful":missing valucs incostsensitive decision trees. IEEE Transactions on Knowledge and Data Engineering, Vol.17 No.12.2005.
[39]Pearson R. K. The Problem of Disguised Missing Data ACM SIGKDD Explorations Newsletter archive Volume 8, Issue 1, June.2006.
[40]Rubin, D. B.Multiple Imputation for Nonresponse in Surveys, Wiley:New York. 1987.
[41]SL. Lauritzen. The EM Algorithm for Graphical Association models with Mssing Data Computational Statistics and Data Analysis,1995:191-201.
[42]Dempster, A. P., Laird, N. M. Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Vol.39,1977:1-38.
[43]N. Friedman, D, Geiger&M. Goldszmidt. Bayesian network classifiers. Machine Learning,1997,29(2-3):131-163.
[44]Geman S., Geman D. Stochastic relaxation, Gibbs distributionand the Bayesia restoration of images. IEEE Trans onPattern Analysis and Machine Intelligence,1984, 6:721-742.
[45]Meng, X.L., Rubin, D. B. Maximum likelihood estimation via the ECM algorithm A general framework. Biometrika,1993,80:267.
[46]Sebastian, P., Ramoni, M. Bayesian inference with missing data using bound an collapse. Journal of Computational and Graphical Statistics,2000,9(4):779.
[47]Joseph G. Ibrahim, Hongtu Zhu, Niansheng Tang. Model Selection Criteria for Missing-Data Problems Using the EM Algorithm.J Am Stat Assoc.2008 December 1; 103 (484):1648-1658.
[48]Schafer J., Graham J. Missing data:Our view of the state of the art. Psychological Methods,2002:147-177.
[49]Roth P. Missing data:A conceptual review for applied psychologists. Personnel Psychology,47.1994:537-560.
[50]Sarle W. S. "Prediction with Missing Inputs, "in Wang, P. P. (ed.), JCIS'98 Proceedings, Vol II, Research Triangle Park, NC,1998:399-402.
[51]Allison P. Missing Data. Sage Publication, Inc,2001.
[52]Wang S. C., Yuan S. M. Research on learing bayesian networks structure with missing data, Journal of Software, vol.15 2004:1024-1048.
[53]Wei W., Ying T. A generic Neural Network Approach for Filling Missing Data in Data Ming, Proceedings of the IEEE International Conference on Systems, Man and Cybernetics,2003.vol.1:862-867.
[54]Jonsson P., Claes W. An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[55]VapnikV N.Statistical Learning Theory. Wiley, New York,1998.
[56]Wang, Q. D et al. A rough set based clustering algorithm and the information theoretical approach to refine clusters. ICMM 2004.
[57]雷蕾,吴乃君,刘鹏,刘兰娟灵敏度分析：分类器中的缺失数据.管理学报.2005(9)：153-156.
[58]Cios KJ, Kurgan L. Trends in Data Mining and Knowledge Discovery. Advanced Techniques in Knowledge Discovery and Data Mining. Springer London.2005:1-26
[59]Marvin L Brown, John F Kros. Data Mining and The Impact of Missing Data Data Mining:Opportunities and Challenges, Idea Group Publishing, USA,2003:611-621.
[60]Quinlan, J. R. Unknown attributes values. In Programs for Machine Learning. Morgan Kaufmann,1993:27-32.
[61]Wang, S. C., and Yuan, S. M. Research on learing bayesian networks structure with missing data, Journal of Software, vol.15.2004:1024-1048.
[62]Jonsson P., Claes W.An evaluation of k-nearest neighbour imputation using likert data, Proceedings-10th international symposium on software metrics,2004.
[63]Shahshahani B, Landgrebe D. The effect of unlabeled samples in reducing the small sample size problem and mitigating the Hughes phenomenon. IEEE Transactions of Geoscience and Remote Sensing,1994,32(5):1087-1095.
[64]0 Chapelle, B Scholkopf, A Zien. Semi-supervised learning. Cambridge MA: MITPress,2006.
[65]Avrim Blum, Tom Mitchell. Combining labeled and unlabeled data with co-training. Proceedings of the eleventh annual conference on Computational learning theory, NewYork:ACMPress,1998:92-100.
[66]Yarowsky D.Unsupervised word sense disambiguation rivaling supervised methods. Proceedings of the 33rd Annual Meeting of the Association for Computatonal Linguistics,1995:189-196.
[67]Riloff E, Wiebe J, Wilson T. Learning subjective nouns using extraction pattern boot strapping. Proceedings of the Seventh Conference on Natural Language Learning (CoNLL-2003),2003.
[68]Zhou Z H, Zhan D C, Yang Q. Semi-supervised learning with very few labeled training examples. Association for the Advancement of Artificial Intelligence.http: //www. aaai. org.
[69]Nigam K, McCallum, A K, Thrun s, et al.Text classification from labeled and unlabeled documents using EM. Machine Learning,2000(39):103-134.
[70]Nigam K. Using unlabeled data to improve text classification (Technical Report CMU-CS-01-126). Carnegie Mellon University, Doctoral Dissertation,2001.
[71]Baluja S. Probabilistic modeling for face orientation discrimination:leaming from labeled and unlabeled data. Neural Information Proeessing Systems,1998.
[72]Fujino A, Ueda N, Saito K.A hybrid generative/discriminative approach to semi-supervised classifier design. The Twentieth National Conference on Artificial Intelligence(AAAI05),2005.
[73]Joachims T. Transductive inference for text classification using support vector maehines. Proeeedings of the 16th International Conference on Machine Learning, San Fransiseo,1999:200-209.
[74]Nigam K, Ghani R. Analyzing the effectiveness and applicability of co-training. Proceedings of the 9th ACM International Conference on Information and Knowledge Management (CIKM2000),Mclean, VA,2000:86-93.
[75]Goldman S, Zhou Y. Enhancing supervised learning with unlabeled data. Proceedings of the 17th International Conference on Machine Learning (ICML2000). San Francisco:Morgan Kaufmann Publishers,2000:327-334.
[76]Zhou Z H, Li M. Tri-training:Exploiting unlabeled data using threclassifiers. IEEE Transactions on Knowledge and Data Engineering,2005,17 (11):1529-1541.
[77]J. Zhu.Semi-Supervised Learning Literature Survey.Computer Sciences, University of Wisconsin-Madison, Technical report 1530,2008.
[78]X. Zhu.Semi-supervised learning with graphs. Doctoral dissertation (CMU-LTI-05-192), Carnegie Mellon University.2005.
[79]Z. Y. Niu, D. H. Ji, C. L. Tan. Word sense disambiguation using label propagation based semi-supervised learning. In Proceedings of the ACL05.2005:395-402.
[80]A.B.Goldberg, X.Zhu. Seeing stars when there aren't many stars:Graph-based semi-supervised learning for sentiment categorization. In HLTNAACL 2006 Workshop on Textgraphs:Graph-based Algorithms for Natural Language Processing. NY, USA.2006. LNCS5012:1006-1014.
[81]L.Grady, G. Funka-Lea. Multi-label image segmentation for medical applications based on graph-theoretic electrical potentials. In ECCV 2004 workshop.17. Cambridge, MA:MIT Press.2004:230-245.
[82]BelkinM, Matveeval, NiyogiP. Regression and regularization on large graphs. Proeeedings of the 17th Annual Conference on Learning Theory, NewYork:ACM Press, 2004:185-192.
[83]P. Domingos, M. Pazzani. On the Optimality of the Simple Bayesian Classifier under Zero-One Loss. Machine Leaming,1997,29(2-3):103-130.
[84]M.J. Pazzani. Searching for Dependencies in Bayesian Classifiers. Learning from Data, Artificial Intelligence and Statiscics. New York, Springer Verlag,1996:239-248.
[85]Zi jian Zheng. Naive bayesian classifier committees. In Proceedings of ECML'98, Berlin:Springer Verlag,1998:196-207.
[86]Zhu XJ. Semi-supervised learning with graphs. USA:Carnegie Mellon University. 2006:1-89.
[87]陈锦秀,姬东鸿.基于图的半监督关系抽取.软件学报.2008,19(11)：2843-2852.
[88]X Zhu, Z Ghahramani, J Lafferty. Semi-supervised learning using Gaussian fields and harmonic functions. Proeeedings of the 20th International conference on Machine Learning (ICML2003), Washington DC,2003:912-919.
[89]D Zhou,O Bousquet, TN Lal, J Weston. Learning, with local and global consisteney. Advances in Neural Information Proeessing Systems16, Cambridge, MA:MIT Press,2004:321-328.
[90]M Belkin, P Niyogi, V Sindwani. On manifold regularization. Proeeedings of the 10th International Workshop on Artificial Intelligence and Statistics(AISTATS2005), 2005:17-24.
[91]Chapelle O, Chi M, Zien A. A continuation method for semi-supervised SVMs.23rd International Conference on Machine Learning. USA, Pittsburgh,2006:1148-1157.
[92]Chapelle O, Scholkopf B, Zien A. Semi-Supervised Learning. Massachusetts.MIT Press,2006:33-543.
[93]Chapelle O,Zien A.Semi-supervised classification by low density separation Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics. Barbados.2005:57-64.
[94]Sindhwani V,Keerthi S, Chapelle 0. Deterministic annealing for semi-supervised kernelmachines. Proceedings of the 23rd International Conference on Machine Learning. USA:Association for Computing Machinery.2006:841-848.
[95]Chapelle 0, Sindhwani V, Keerthi SS. Branch and bound for semi-supervised support vectormachines. Advances in Neural Information Processing Systems, USA:MIT Press.2006:217-224.
[96]陈毅松.基于支持向量机的渐进直推式分类学习算法.软件学报,2003,3：459-460.
[97]Reshmi Malhotra, DK Malhotra. Differentiating between good credits and bad credits using neuro-fuzzy systems. Computing artificial intelligence and information technology.2002,136(1):190-211.
[98]Murphy P M, Aha D W. UCI Machine Learning Repository,2009-1-9. http://www. ics. uci. edu/～mlearn.
[99]WEKA. http://www. cs. waikato.ac.nz/ml/weka/.2009-01-11.
[100]Giorgio Corani, Marco Zaffalon. JNCC2:The Java Implementation Of Naive Credal Classifier 2. Journal of Machine Learning Research.2008,9:2695-2698.
[101]Lann. Stylometry and method. The case of Euripides. Literary and linguistic computing,1995, (10):271-278.
[102]Baayen, Harald, Hans, van Halteren & Fiona Tweedie. Outside the Cave of Shadows: Using Syntactic Annotation to Enhance Authorship Attribution. Literary and Linguistic Computing,1996,11(3):121-131.
[103]Stamatos, Efstathios, Nikos Fakotakis, George Kokkinakis. Automatic Text Categorization in Terms of Genre and Author. Computational Linguistics,2000,26(4): 471-495.
[104]Michael Gamon. Linguistic correlates of style:authorship classification with deep linguistic analysis features. In Proceeding 20th International Conference of Computational Linguistics, Geneva,2004:611-617.
[105]霍跃红,典籍英译译者文体分析与译者文本识别.博士学位论文.大连：大连理工大学,2010.
[106]金奕江,孙晓明,马少平.因特网上的写作风格鉴别.广西师范大学学报(自然科学版).2003,21(1)：62-66.
[107]马建斌.基于SVM的中文电子邮件作者身份挖掘技术研究.硕十学位论文.保定：河北农业大学,2004.
[108]常淑惠.基于写作风格的中文邮件作者身份识别技术研究.硕士学位论文.保定：河北农业大学,2005.
[109]武晓春,黄萱菁,吴立德.基于语义分析的作者身份识别方法研究.中文信息学报.2006,20(6)：61-68.
[110]Mat sura, Kanada. Authorship Detection of Sentences by 8 Japanese Modern Authors via N-gram Distribution. IPSJ SIG Notes,2000-NL-137:1-8.
[111]Yoshida, Nobesawa, Saito. Effective Features of Authorship Identification. IPSJ SIG Notes,2001-NL-145:83-90.
[112]J. Hoorn, S. Frank, W. Kowalczyk, F. van der Ham. Neural netword indentif ication of poets using letter sequences. Literary and Linguistic Computing,14(3):311-338.
[113]Cagatay Catal, Kemalettin Erbakici, Yasar Erenler. Computer-Based authorship attribution for Turkish documents. IJCI proceeding of Intl. Ⅻ. Tukish Symoposium on Artificail Interlligence and Neural Networks.2003.
[114]Zhao Ying, Zobel Justin. Effective and Scalable Authorship Attribution Using Function Words. Lecture Notes in Computer Science.2005:174-189.
[115]易勇.计算机辅助诗词创作中的风格辨析及联语应对研究.博士学位论文.重庆：重庆大学,2005.
[116]张全,张运良,袁毅.利用语言概念表示的作者写作风格分类研究.第七届中文信息处理国际会议论文集.2007：460-464.
[117]张运良,朱礼军,乔晓东,张全.基于句类特征的作者写作风格分类研究.计算机工程与应用.2009.22
[118]吴春龙周昌乐,基于频繁关键字共现的诗词风格分类模型研究,厦门大学学报(自然科学版).2008(1)：41-44.
[119]Miller D J, Uyar H S. A mixture of experts classifier with learning based on both labelled and unlabelled data. Mozer M, Jordan M I, Petsche T, Advances in Neural Information Processing Systems 9, Cambridge, MA:MIT Press,1997:571-577.
[120]T.Zhang, F. J. Oles.A probability analysis on the value of unlabeled data for classification problems. In:Proceedings of the 17th International Conference on Machine Learning (ICML'00), San Francisco, CA,2000:1191-1198.
[121]0. Chapelle and B. Schokopf and A. Zien. Semi-Supervised Learning. MIT Press, Cambridge, MA,2006.
[122]周志华,王珏.机器学习及其应用.北京：清华大学出版社,2007.
[123]Xiaojin Zhu. Semi-supervised learning literature survey. TR-1530. University of Wisconsin-Madison Department of Computer Science.2005.
[124]Xiaojin Zhu. Semi-Supervised Learning Tutorial. Tutorial of ICML07,2007.
[125]W. Wang, Z.-H. Zhou. Analyzing co-training style algorithms. In:Proceedings of the 18th European Conference on Machine Learning (ECML'07), Warsaw, Poland,2007.
[126]Zhong S. Semi-Supervised model-based document clustering:A comparative study. Machine Learning,2006,65(1):3-29.
[127]Bouchachia A, Pedrycz W. Data clustering with partial supervision. Data Mining and Knowledge Discovery,2006,12(1):47-78.
[128]邓超,郭茂祖.基于自适应数据剪辑策略的Tri-training算法.计算机学报.2007(9)：1213-1226.
[129]Guifa Teng, Yihong Liu, Jianbin Ma, Fang Wang, Huiting Yao. Improved Algorithm for Text Classification Based on TSVM. Proceedings of the First International Conference on Innovative Computing, Information and Control.2006:55-58.
[130]YE WANG, Shang-Teng HUANG.Training TSVM with the proper number of positive samples. Pattern recognition letters. Elsevier,2005:1414,2187-2194.
[131]廖东平,姜斌,魏玺章,黎湘,庄钊文.一种快速的渐进直推式支持向量机分类学习算法.系统工程与电子技术,2007,29(1)：87-91.
[132]Ramesh Agarwal, Mahesh V. Joshi. IBM ResearchReport. PNrule:A new Framework for Learning Classifier Models in Data Mining (A Case-Study in Network Intrusion Detection). In Proceedings of First SIAM Conference on DataMining, Chieago, April,2001.
[133]Goldman S, Zhou Y. Enhancing supervised learning with unlabeled data. In Proceedings of 17th International Conference on Machine Learning (ICMLOO),2000, 327-334.
[134]Zhou Y, Goldman S. Democratic co-learing.In Proceedings of thel6th IEEE International Conference on Tools with Artificial Intelligence,2004.
[135]郝秀兰,胡运发.半监督的文本分类——两阶段协同学习.小型微型计算机系统.Vo1.30 No.10,2009：1921-1924.
[136]Giorgio Corani, Marco Zaffalon. JNCC2:The Java Implementation of Naive Credal Classifier 2. Journal of Machine Learning Research.2008,9:2695-2698.
[137]Joachims. Multi-class Support Vector Machine. http://www.cs.cornell.edu/ People/tj/svm_light/svm_multiclass. html.2008.8.
[138]Mccallum. A Mccallum Rainbow. http://www. cs. cmu. edu/~mccallum/bow/ rainbow.2008.9.
[139]WALLEY P. Statistical Reasoning with Imprecise Probabilities. London:Chapman and Hall.1991.
[140]IGOR 0 K, YEVGENY V F. Imprecise system reliability. International Journal of Systems Science,2001,32 (4):487-493.
[141]Weichselberger K. The theory of interval-probability as a unifying concept for uncertainty.lnernational Journal of Approximate Reasonging,2000, Vol.24:149-170.
[172]冯向前.区间数不确定多属性决策方法研究.博士学位论文.南京：南京航空航天大学,2007.