基于正例和无标记样例学习研究
|
摘要
随着World Wide Web的迅猛发展,Web信息呈现出爆炸式指数级涌现,催生了搜索引擎这一激动人心的研究领域。各种搜索引擎已经成为人们使用因特网的最重要的信息服务工具,但是人们要想找到许多重要信息仍然如同大海捞针一般,研究者们公认面向主题的搜索是搜索引擎未来最重要的发展方向。主题爬行(Focused Crawling)系统采用基于样例网页驱动的主题信息收集方法,具有重要的学术研究价值和广阔的应用前景。
本文即是针对主题爬行中的关键技术——文本分类问题,将主题相关性转变为基于正例和无标记样例的学习(Learning from Positive and Unlabeled examples,PU学习)问题。PU学习范型的最大问题是没有可以利用的反例,因此传统的监督学习和半监督学习方法不能有效的使用。本文针对这一学习范型进行了跟踪,做了比较全面的综述和深入的研究,将基于机器学习的文本挖掘技术引入PU学习,并加以应用,提出了新颖的解决办法,取得丰硕而有成效的研究成果。本文创新工作主要包括如下两个方面:
第一方面工作是基于两阶段策略的研究工作,针对两阶段策略中的第一阶段——提取可靠反例,提出了三种有效的可靠反例提取算法:(1)基于经典的k-Means聚类算法的可靠反例提取算法,首先对训练集合(正例集合和无标记样例集合)采用k-Means聚类算法进行聚类,将正例比例低于某一阈值的簇标记为可靠反例;(2)基于约束k-Means聚类的可靠反例提取算法,约束k-Means聚类是一种全新的半监督聚类算法,在聚类过程中用正例集合来初始化正例中心,将正例标记做为Must-link约束进行约束聚类,本方法最后不仅标记了可靠反例,也同时扩充了正例集合;(3)基于kNN的Ranking学习算法的可靠反例提取算法,将无标记样例采用kNN算法计算其与k个正例近邻的Rank值,将Rank值低于一定阈值的样例标记为可靠反例。
第二方面工作是基于协同训练范型这一半监督学习中最重要的方法提出了两种PU学习算法:(1)基于Co-EM SVM的PU学习,Co-EM SVM是对标准协同训练算法在EM算法框架之下使用SVM做为内嵌分类器的改进。首先采用基于1-DNF方法的视图划分方法,将文本特征集合划分为正例特征集和反例特征集组成两个视图,然后在单视图上提取可靠反例,最后采用Co-EM SVM进行迭代学习。(2)基于Tri-training算法的PU学习,Tri-training是采用单视图多分类器方法对协同训练算法的推广,本文采用了三个已有的可靠反例提取算法分别初始化三个SVM分类器,然后将其两个分类器的一致分类结果作为第三个分类器的训练样例进行迭代学习,最终分类结果通过三个分类器的集成得到。
本文提出的方法均在经典的文本分类数据集上与相关工作进行了对比实验,并采用通常的文本分类评估指标,验证了本文工作明显优于相关工作,取得了较好的实验效果,并就本文工作进行了总结,公开发表了相关的学术论文,取得了较好的评价。
With the rapid development of World Wide Web, the emergence of information in the Web presents an explosive and exponential way, and searching for something that people required likes looking for a needle in a haystack, and this is the so-called "information explosion" problem, that is, information is greatly abundant while the knowledge is relatively scarce. Thus came the birth of the search engine, which has turned out an exciting area for research. Now it has become one of most important applications of Internet information service tools. However, due to the nature of its free structure and constant dynamic change, sometimes people can not access the important information that they need, and focused crawling is the basic approach to solving it.
The main method of focused crawling comes from the system that S. Chakrabarti built in 1999, which adopted the method for collecting the focused information based on examples web page driving. The technical difficulty of focused crawling lies in whether the users can make an accurate anticipation of the related topic of the web page before downloading the actual page, which can come true only through the machine learning techniques. Funded by the National Natural Science Foundation "Based on the Incremental and Mobile Characteristics of the Focused Crawling Technology" (Grant No. 60373099), this thesis studies the key technology of focused crawling-text classification, and regards the prediction of topic relevance as the relevance between documents, thereby the judgment of topics is changed to an issue of learning from positive and unlabeled examples (LPU), and a summary is made of the attempt and exploration of it.
The biggest problem of LPU is lack of negative examples that can be used, so the traditional algorithms of supervised learning and semi-supervised learning can not be effectively applied. This thesis firstly traces the paradigm of LPU,and makes comprehensive surveys of the related work and an in-depth study. The text mining technologies based on the up-to-date machine learning are introduced into the LPU issue and then get disseminated. This thesis puts forward some new solutions and attains some fruitful and productive research results. The main innovation of this thesis includes two aspects:
The first aspect of the works is based on the two-stage strategy, which is the most studied problem: in stage 1, extracting a set of negative examples called reliable negatives (RN) from the unlabeled examples U as the initial negative examples; in stage 2, building a set of classifiers by iteratively applying a classification algorithm for U minus RN set, and expanding the quantity of reliable negative examples, thus the final classifier is obtained. Since Support Vector Machine (SVM) has been confirmed as one of the best classifiers in text classification, so usually the iterative Support Vector Machine method is used in stage 2. The research is mainly concentrated in stage 1--extracting reliable negative examples. Based on the machine learning technology, this thesis puts forward three novel algorithms for reliable negative examples extraction:
(1)A novel algorithm for extracting reliable negative examples has been put forward based on the classical k-Means clustering algorithm. Clustering assumption, a basic assumption for semi-supervised learning, has been used widely in the semi-supervised learning. Text classification method aided by clustering can be divided into three major study fields, that is, feature selection based on clustering, clustering in semi-supervised classification, clustering in large-scale classification problems. The proposed method firstly uses k-Means algorithm to cluster the training set (including the positive examples set and unlabeled examples set), and then labels some cluster where the proportion of positive examples is below a certain threshold as reliable negative examples. Experiments show that the quantity of reliable negative examples obtained in this method is moderate, and well balances the ratio of the initial positive and negative examples, and with a strong ability to distinguish, so good experiment effect is obtained.
(2)Another novel algorithm of extracting reliable negative examples has been put forward based on the constraint-based k-Means clustering. The existing small amount of supervised information is used to boost the clustering performance in the classical clustering algorithm, and thus has given rise to the semi-supervised clustering algorithm, one of the most important studies is constraint-based clustering. In order to use the existing positive examples label in the clustering algorithm, the constrained k-Means clustering algorithm is introduced, which is a relatively new semi-supervised clustering algorithm. In clustering performance, firstly, the positive examples set is used to initialize the centroid, and the positive examples label is used as the Must-link constraint. And then some proportion of examples near the centroid is labeled. It should be noted that this method also expands some positive examples. This work is still rare in current study at LPU, and is only used in PNLH algorithm, it is an innovative attempt in this thesis.
(3) The third novel algorithm of extracting reliable negative examples has been put forward based on the Ranking learning using kNN algorithm. Ranking learning originates from the field of information retrieval, and is a machine learning method between classification and regression. kNN algorithm has been used to sort the unlabeled examples by computing the similarity of unlabeled examples and its positive k-nearest neighbor. Instead of using the usual method that labels the examples that have the highest rank, but labeling the examples with a lower threshold value as reliable negative examples, and discuss threshold selection issues through experiments.
Two novel LPU methods have been proposed as the second aspect work in this thesis based on Co-training paradigm, which is the most important semi-supervised learning method. Standard Co-Training paradigm assumes that the attribute set has two sufficient and redundant view, that is, two attribute sets must meet the conditional independence assumptions: (1) Each attribute set is sufficient to describe the problem, in other words, each attribute set can be sufficient to train a strong learner if it has enough training sample; (2) given the label, each attribute set is conditionally independent of another attribute set. Thus a separate classifier can be trained for each view using the labeled examples, then, in Co-training performance, each classifier selects and labels a number of examples with higher confidence, and updates another classifier with the newly labeled examples. Co-training algorithm runs a continuously iterative process until meets a stop condition. As a typical semi-supervised learning method, the Co-training algorithm searches the unlabeled examples feature space with complementary classifier. In general, the complementary classifier can be attained by characteristics or complementary model. The effectively intuitive explanation of this approach is: the uncertain results of one classifier may be identified for complementary classifier. The final results are complementary to each other through integrated classifier results. In the case of an appropriate classifier, Co-training algorithm performs better than self-training algorithm. This thesis puts forward two new algorithms based on two most important Co-training algorithms: the Co-training and the Tri-training algorithm:
(1)LPU based on Co-EM SVM algorithm. Co-EM SVM algorithm has been improved from the standard Co-training algorithm under EM algorithm framework and replaced the original embedded NB classifier with SVM classifier, which has been used commonly in the field of text classifier, and also been proved to be one of the best classifiers. The proposed method firstly splits the views based on 1-DNF algorithm into the positive feature set and negative feature set, and then runs a reliable negative extraction algorithm in a single view to extract the reliable negative examples for the view, thus ensures the training of the initial classifier, and finally Co-EM SVM algorithm can be applied to LPU, with better results through experiments.
(2)LPU based on Tri-training algorithm. Tri-training algorithm further relaxes the constraints of Co-training algorithm, and does not require sufficient and redundant view, nor does it need to use different types of classifiers. The notable feature of Tri-training algorithm is the usage of three classifiers, not only can it easily deal with the estimated confidence level, but also solve the prediction problem of unlabeled examples, and can also use ensemble learning to improve the generalization ability. The proposed method uses the reliable negative examples extracted by the extraction algorithm of the three existing reliable negative examples to initialize the three SVM classifiers, instead of the bootstrap sampling process of standard Tri-training algorithm; Examples labeling strategy is adding the training set to the third classifier when other two classifiers have both determines an example as positive or negative; Termination condition is when three classifiers have no examples to classify; the final output is a combination of the three classifiers. The method is proved effective through experiment.
In this thesis, the proposed work has been compared through experiments in the popular text classification data set-Reuters 21578, under evaluation measures of text classification, and the experiment results show that proposed works are better than related works, and the research results were issued. Although some achievements have been attained in LPU, there is room to improve and go further. For instance, this thesis does not take into account related issues of the manifold assumptions in the semi-supervised learning; how to study the distance function of the semi-supervised clustering as well as kernel-based clustering issues; another issue is the lack of formal verification and extension of the effectiveness of Co-training algorithm, although a view splitting algorithm based on positive examples is put forward, it is only an empirical conclusion and has not given out formal proofs. Since little research is done currently in this work, it is worthy of in-depth research and extension. In short, the research and application of LPU is in its infancy stage, there are great deals of things for our profound research, which will lead the author into further exploration.
本文即是针对主题爬行中的关键技术——文本分类问题,将主题相关性转变为基于正例和无标记样例的学习(Learning from Positive and Unlabeled examples,PU学习)问题。PU学习范型的最大问题是没有可以利用的反例,因此传统的监督学习和半监督学习方法不能有效的使用。本文针对这一学习范型进行了跟踪,做了比较全面的综述和深入的研究,将基于机器学习的文本挖掘技术引入PU学习,并加以应用,提出了新颖的解决办法,取得丰硕而有成效的研究成果。本文创新工作主要包括如下两个方面:
第一方面工作是基于两阶段策略的研究工作,针对两阶段策略中的第一阶段——提取可靠反例,提出了三种有效的可靠反例提取算法:(1)基于经典的k-Means聚类算法的可靠反例提取算法,首先对训练集合(正例集合和无标记样例集合)采用k-Means聚类算法进行聚类,将正例比例低于某一阈值的簇标记为可靠反例;(2)基于约束k-Means聚类的可靠反例提取算法,约束k-Means聚类是一种全新的半监督聚类算法,在聚类过程中用正例集合来初始化正例中心,将正例标记做为Must-link约束进行约束聚类,本方法最后不仅标记了可靠反例,也同时扩充了正例集合;(3)基于kNN的Ranking学习算法的可靠反例提取算法,将无标记样例采用kNN算法计算其与k个正例近邻的Rank值,将Rank值低于一定阈值的样例标记为可靠反例。
第二方面工作是基于协同训练范型这一半监督学习中最重要的方法提出了两种PU学习算法:(1)基于Co-EM SVM的PU学习,Co-EM SVM是对标准协同训练算法在EM算法框架之下使用SVM做为内嵌分类器的改进。首先采用基于1-DNF方法的视图划分方法,将文本特征集合划分为正例特征集和反例特征集组成两个视图,然后在单视图上提取可靠反例,最后采用Co-EM SVM进行迭代学习。(2)基于Tri-training算法的PU学习,Tri-training是采用单视图多分类器方法对协同训练算法的推广,本文采用了三个已有的可靠反例提取算法分别初始化三个SVM分类器,然后将其两个分类器的一致分类结果作为第三个分类器的训练样例进行迭代学习,最终分类结果通过三个分类器的集成得到。
本文提出的方法均在经典的文本分类数据集上与相关工作进行了对比实验,并采用通常的文本分类评估指标,验证了本文工作明显优于相关工作,取得了较好的实验效果,并就本文工作进行了总结,公开发表了相关的学术论文,取得了较好的评价。
With the rapid development of World Wide Web, the emergence of information in the Web presents an explosive and exponential way, and searching for something that people required likes looking for a needle in a haystack, and this is the so-called "information explosion" problem, that is, information is greatly abundant while the knowledge is relatively scarce. Thus came the birth of the search engine, which has turned out an exciting area for research. Now it has become one of most important applications of Internet information service tools. However, due to the nature of its free structure and constant dynamic change, sometimes people can not access the important information that they need, and focused crawling is the basic approach to solving it.
The main method of focused crawling comes from the system that S. Chakrabarti built in 1999, which adopted the method for collecting the focused information based on examples web page driving. The technical difficulty of focused crawling lies in whether the users can make an accurate anticipation of the related topic of the web page before downloading the actual page, which can come true only through the machine learning techniques. Funded by the National Natural Science Foundation "Based on the Incremental and Mobile Characteristics of the Focused Crawling Technology" (Grant No. 60373099), this thesis studies the key technology of focused crawling-text classification, and regards the prediction of topic relevance as the relevance between documents, thereby the judgment of topics is changed to an issue of learning from positive and unlabeled examples (LPU), and a summary is made of the attempt and exploration of it.
The biggest problem of LPU is lack of negative examples that can be used, so the traditional algorithms of supervised learning and semi-supervised learning can not be effectively applied. This thesis firstly traces the paradigm of LPU,and makes comprehensive surveys of the related work and an in-depth study. The text mining technologies based on the up-to-date machine learning are introduced into the LPU issue and then get disseminated. This thesis puts forward some new solutions and attains some fruitful and productive research results. The main innovation of this thesis includes two aspects:
The first aspect of the works is based on the two-stage strategy, which is the most studied problem: in stage 1, extracting a set of negative examples called reliable negatives (RN) from the unlabeled examples U as the initial negative examples; in stage 2, building a set of classifiers by iteratively applying a classification algorithm for U minus RN set, and expanding the quantity of reliable negative examples, thus the final classifier is obtained. Since Support Vector Machine (SVM) has been confirmed as one of the best classifiers in text classification, so usually the iterative Support Vector Machine method is used in stage 2. The research is mainly concentrated in stage 1--extracting reliable negative examples. Based on the machine learning technology, this thesis puts forward three novel algorithms for reliable negative examples extraction:
(1)A novel algorithm for extracting reliable negative examples has been put forward based on the classical k-Means clustering algorithm. Clustering assumption, a basic assumption for semi-supervised learning, has been used widely in the semi-supervised learning. Text classification method aided by clustering can be divided into three major study fields, that is, feature selection based on clustering, clustering in semi-supervised classification, clustering in large-scale classification problems. The proposed method firstly uses k-Means algorithm to cluster the training set (including the positive examples set and unlabeled examples set), and then labels some cluster where the proportion of positive examples is below a certain threshold as reliable negative examples. Experiments show that the quantity of reliable negative examples obtained in this method is moderate, and well balances the ratio of the initial positive and negative examples, and with a strong ability to distinguish, so good experiment effect is obtained.
(2)Another novel algorithm of extracting reliable negative examples has been put forward based on the constraint-based k-Means clustering. The existing small amount of supervised information is used to boost the clustering performance in the classical clustering algorithm, and thus has given rise to the semi-supervised clustering algorithm, one of the most important studies is constraint-based clustering. In order to use the existing positive examples label in the clustering algorithm, the constrained k-Means clustering algorithm is introduced, which is a relatively new semi-supervised clustering algorithm. In clustering performance, firstly, the positive examples set is used to initialize the centroid, and the positive examples label is used as the Must-link constraint. And then some proportion of examples near the centroid is labeled. It should be noted that this method also expands some positive examples. This work is still rare in current study at LPU, and is only used in PNLH algorithm, it is an innovative attempt in this thesis.
(3) The third novel algorithm of extracting reliable negative examples has been put forward based on the Ranking learning using kNN algorithm. Ranking learning originates from the field of information retrieval, and is a machine learning method between classification and regression. kNN algorithm has been used to sort the unlabeled examples by computing the similarity of unlabeled examples and its positive k-nearest neighbor. Instead of using the usual method that labels the examples that have the highest rank, but labeling the examples with a lower threshold value as reliable negative examples, and discuss threshold selection issues through experiments.
Two novel LPU methods have been proposed as the second aspect work in this thesis based on Co-training paradigm, which is the most important semi-supervised learning method. Standard Co-Training paradigm assumes that the attribute set has two sufficient and redundant view, that is, two attribute sets must meet the conditional independence assumptions: (1) Each attribute set is sufficient to describe the problem, in other words, each attribute set can be sufficient to train a strong learner if it has enough training sample; (2) given the label, each attribute set is conditionally independent of another attribute set. Thus a separate classifier can be trained for each view using the labeled examples, then, in Co-training performance, each classifier selects and labels a number of examples with higher confidence, and updates another classifier with the newly labeled examples. Co-training algorithm runs a continuously iterative process until meets a stop condition. As a typical semi-supervised learning method, the Co-training algorithm searches the unlabeled examples feature space with complementary classifier. In general, the complementary classifier can be attained by characteristics or complementary model. The effectively intuitive explanation of this approach is: the uncertain results of one classifier may be identified for complementary classifier. The final results are complementary to each other through integrated classifier results. In the case of an appropriate classifier, Co-training algorithm performs better than self-training algorithm. This thesis puts forward two new algorithms based on two most important Co-training algorithms: the Co-training and the Tri-training algorithm:
(1)LPU based on Co-EM SVM algorithm. Co-EM SVM algorithm has been improved from the standard Co-training algorithm under EM algorithm framework and replaced the original embedded NB classifier with SVM classifier, which has been used commonly in the field of text classifier, and also been proved to be one of the best classifiers. The proposed method firstly splits the views based on 1-DNF algorithm into the positive feature set and negative feature set, and then runs a reliable negative extraction algorithm in a single view to extract the reliable negative examples for the view, thus ensures the training of the initial classifier, and finally Co-EM SVM algorithm can be applied to LPU, with better results through experiments.
(2)LPU based on Tri-training algorithm. Tri-training algorithm further relaxes the constraints of Co-training algorithm, and does not require sufficient and redundant view, nor does it need to use different types of classifiers. The notable feature of Tri-training algorithm is the usage of three classifiers, not only can it easily deal with the estimated confidence level, but also solve the prediction problem of unlabeled examples, and can also use ensemble learning to improve the generalization ability. The proposed method uses the reliable negative examples extracted by the extraction algorithm of the three existing reliable negative examples to initialize the three SVM classifiers, instead of the bootstrap sampling process of standard Tri-training algorithm; Examples labeling strategy is adding the training set to the third classifier when other two classifiers have both determines an example as positive or negative; Termination condition is when three classifiers have no examples to classify; the final output is a combination of the three classifiers. The method is proved effective through experiment.
In this thesis, the proposed work has been compared through experiments in the popular text classification data set-Reuters 21578, under evaluation measures of text classification, and the experiment results show that proposed works are better than related works, and the research results were issued. Although some achievements have been attained in LPU, there is room to improve and go further. For instance, this thesis does not take into account related issues of the manifold assumptions in the semi-supervised learning; how to study the distance function of the semi-supervised clustering as well as kernel-based clustering issues; another issue is the lack of formal verification and extension of the effectiveness of Co-training algorithm, although a view splitting algorithm based on positive examples is put forward, it is only an empirical conclusion and has not given out formal proofs. Since little research is done currently in this work, it is worthy of in-depth research and extension. In short, the research and application of LPU is in its infancy stage, there are great deals of things for our profound research, which will lead the author into further exploration.
引文
1 http://www.netcraft.com/archives/web_server_survey.html
2 https://sites.google.com/site/googol/Home/google-news/2008-%E5%B9%B4-7-%E6%9C%88
1 http://www.cnnic.net.cn/uploadfiles/doc/2009/1/13/92209.doc
[1] Soumen Chakrabarti, M. van den Berg and B. Dom.Focused Crawling: A New Approach to Topic-Specific Web Resource Discovery [J]. Computer Networks, 1999, 31(11-16): 1623–1640.
[2] B. Liu. Web Data Mining Exploring Hyperlinks, Contents and Usage Data [M], Springer, December, 2006.
[3]李晓明,闫宏飞,王继民.搜索引擎-原理、技术与系统[M].北京:科学出版社,2004.
[4] S. Brin and L. Page.The Anatomy of a Large-scale Hypertextual Web Search Engine [J]. Computer Networks, 1998, 30(1-7): 107–117.
[5]王珏,周志华,周傲英.机器学习及其应用[C].北京:清华大学出版社,2006.
[6] Soumen Chakrabarti . Mining the Web Discovering Knowledge from Hypertext Data [M].Morgan-Kaufmann Publishers, 2002.
[7] Fabrizio Sebastiani. A Tutorial on Automated Text Categorization [C], In Analia Amandi and Alejandro Zunino (eds.), Proceedings of ASAI-99, 1st Argentinian Symposium on Artificial Intelligence, Buenos Aires, AR, 1999, 7-35.
[8] J. M. Kleinberg. Authoritative Sources in a Hyperlinked Environment [J], J. ACM, 1999, 46(5): 604–632.
[9] B. D. Davison. Topical Locality in the Web [C], SIGIR2000, 272–279, 2000.
[10] S. Chakrabarti, K. Punera and M. Subramanyam. Accelerated Focused Crawling through Online Relevance Feedback [C], WWW2002, 148–159, 2000.
[11] P. D. Bra, G. Houben, Y. Kornatzky, and R. Post. Information Retrieval in Distributed Hypertexts [C], Procs. of the 4th RIAO Conference, New York, 481–491, 1994.
[12] P D Bra, et al. Searching for Arbitrary Information in the WWW: The Fish-Search for Mosac [C], WWW Conference, 1994.
[13] M. Hersovici, M. Jacovi, Y. S. Maarek, D. Pellegb, M. Shtalhaima, and S. Ura. The Shark-Search Algorithm, An Application: Tailored Web Site Mapping [C]. In WWW7. 1998.
[14] J. Cho, H. Efficient Crawling Through URL Ordering [C], Garcia-Molina, L. Page. In Proceedings of the 7th International WWW Conference, Brisbane, Australia, April 1998.
[15] L. Page, S. Brin, R. Motwani, T. Winograd. The PageRank Citation Ranking: Bringing Order to the Web [R], Stanford Digital Library Technologies Project, 1998.
[16] Menczer F,Pant G,Ruiz M,Srinivasan P. Evaluating Topic-Driven Web Crawlers [C]. In: 24th Annual Intl. ACM SIGIR Conf. on Research and Development in Information Retrieval, 2001, New York, 241-249.
[17] S. Chakrabarti, David A. Gibson, and Kevin S. Mccurley. Surfing the Web Backwards [C], In Proceedings of the 8th International WWW Conference. Toronto, Canada, 1679-1693 May 1999.
[18] A. McCallum, K. Nigam, J. Rennie, and K. Seymore. Building Domain-Specific Search Engines with Machine Learning Technique [C], In Procs. of AAAI Spring Symposium on Intelligents Engine in Cyberspace, 1999.
[19] Rennie J.,McCallum A.Using Reinforcement Learning to Spider the Web Efficiently [C], In:Proceedings of ICML-99, 16th International Conference on Machine Learning, Bled, Slovenia, 1999, 335-343.
[20] M. Diligenti, F. M. Coetzee, S. Lawrence, C. L. Giles, and M. Gori, Focused Crawling using Context Graphs [C], In Procs. of the 26th VLDB Conference, Cairo, Egypt, 2000.
[21] F Menczer, G Gant, P Srinivasan. Topic-Driven Crawlers: Machine Learning Issues [C], ACM TOIT, 2002.
[22] Chiasen Chung, Charles L. A. Clarke. Topic-Oriented Collaborative Crawling [C], CIKM’02, November McLean, Virginia, USA, 2002.
[23] Jan Fiedler, Joachim Hammer. Using the Web Efficiently: Mobile Crawlers[C], Proc of 7th AoM/IaoM Intl Confrence on Computer Science, San Diego, CA, 1999.
[24] M. Ehrig, A. Maedche. Ontology-Focused Crawling of Web Documents [C], Proceedings of the 2003 ACM symposium on Applied computing, 2003.
[25] Silva, I., B. Ribeiro-Neto, P. Calado, N. Ziviani, and E. Moura. Link-based and Content-based Evidential Information in A Belief Network Model [C], Proceedings of the 23rd International ACM SIGIR Conference on Research and Development in Information Retrieval, 96-103, 2000.
[26] Amento, B., L. Terveen, and W. Hill. Does“Authority”Mean Quality? Predicting Expert Quality Ratings of Web Documents[C]. Proc. 23rd ACM SIGIR Conf. on Research and Development in Information Retrieval, 296-303, 2000.
[27] Pant, G., P. Srinivasan, and F. Menczer. Exploration versus Exploitation in Topic Driven Crawlers [C]. In: Proc. WWW-02 Workshop on Web Dynamics, 2002.
[28] Spink, A., D. Wolfram, B. Jansen, and T. Saracevic. Searching the Web: The Public and Their Queries [J]. Journal of the American Society for Information Science, 52(3):226–234, 2001.
[29]黄菁萱,吴立德.基于向量空间模型的文档分类系统[J].模式识别与人工智能, 11(2), 1998
[30] Shih LK, Karger DR. Using URLs and Table Layout for Web Classification Tasks [C]. Proc. of the 13th Int’l Conf. on the World Wide Web (WWW-2004), 193-202, 2004.
[31] Tom M. Mitchell. Machine Learning [M]. McGraw-Hill Companies, Inc, 1997.
[32] O. Chapelle and B. Schokopf and A. Zien. Semi-Supervised Learning [M]. MIT Press, Cambridge, MA,2006.
[33] Jiawei Han, Micheline Kamber著,范明,孟小峰译.数据挖掘概念与技术[M].第二版,北京:机械工业出版社,2007年3月.
[34] Pang-Ning Tan, Michael Steinbach, Vipin Kumar著,范明,范宏建译.数据挖掘导论[M].北京:人民邮电出版社, 2006年5月.
[35]周志华,王珏.机器学习及其应用2007 [M].北京:清华大学出版社, 2007.
[36] B. Shahshahani, D. Landgrebe, The Effect of Unlabeled Samples in Reducing the Small Sample Size Problem and Mitigating the Hughes Phenomenon [J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(5): 1087-1095.
[37] R. P. Lippmann. Pattern Classification using Neural Networks [J]. IEEE Communications, 1989, 27(11): 47-64.
[38] D. J. Miller, H. S. Uyar. A Mixture of Experts Classifier with Learning based on Both Labeled and Unlabeled Data [C]. In: M. Mozer, M. I. Jordan, T. Petsche, eds, Advances in Neural Information Processing Systems 9, Cambridge, MA: MIT Press, 1997, 571-577.
[39] A. Mccallum, K. Nigam. Text Classification by Bootstrapping with Keywords, EM and Shrinkage [C]. Workshop On Unsupervised Learning In Natural Language Processing, 1999, 52-58.
[40] K. Nigam, A. K. McCallum, S. Thrun, T. Mitchell. Using EM to Classify Text from Labeled and Unlabeled Documents [J], Machine Learning, 1998, AAAI Press, 39-103.
[41] K. Nigam, A. K. McCallum, S. Thrun, T. Mitchell. Text Classification from Labeled and Unlabeled Documents using EM [J]. Machine Learning, 2000, 39(2-3): 103-134.
[42] A. Blum, S. Chawla. Learning from Labeled and Unlabeled Data using Graph Mincuts [C]. Proceedings of the 18th International Conference on Machine Learning (ICML’01), San Francisco, CA, 2001, 19-26.
[43] X. Zhu, Z. Ghahramani, J. Lafferty. Semi-Supervised Learning using Gaussian Fields and Harmonic Functions [C] In: Proceedings of the 20th International Conference on Machine Learning (ICML’03), Washington, DC, 2003, 912-919.
[44] M. Belkin, P. Niyogi. Semi-Supervised Learning on Riemannian Manifolds [J]. Machine Learning, 2004, 56(1-3): 209-239.
[45] D. Zhou, O. Bousquet, T. N. Lal, J. Weston, B. Sch?lkopf. Learning with Local and Global Consistency [C]. In: S. Thrun, L. Saul, B. Sch?lkopf, eds. Advances in Neural Information Processing Systems 16, Cambridge, MA: MIT Press, 2004, 321-328.
[46] M. Belkin, P. Niyogi, V. Sindwani. On Manifold Regularization [C]. In: Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics (AISTATS’05), Savannah Hotel, Barbados, 2005, 17-24.
[47] A. Blum, T. Mitchell. Combining Labeled and Unlabeled Data with Co-Training [C]. In: Proceedings of the 11th Annual Conference on Computational Learning Theory (COLT’98), Wisconsin, MI, 1998, 92-100.
[48] Z.-H. Zhou and M. Li. Tri-Training: Exploiting Unlabeled Data using Three Classifiers [J]. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(11): 1529–1541.
[49] F. Denis. PAC Learning from Positive Statistical Queries [C]. The 9th International Workshop on Algorithmic Learning Theory (ALT'98), LNCS, Springer, Heidelberg, 1998, 112-126.
[50] B. Liu, W.S. Lee, P.S. Yu and X.L. Li. Partially Supervised Classification of Text Documents [C]. Proceedings of the Nineteenth International Conference on Machine Learning (ICML-2002), 8-12, July 2002.
[51] Bangzuo Zhang, Wanli Zuo. Learning from Positive and Unlabeled Examples: A Survey [C].2008 International Symposiums on Information Processing and 2008 International Pacific Workshop on Web Mining and Web-based Application,Moscow, Russia, May 23-25, 2008, 640-644.
[52] B. Liu, Y. Dai, X.L. Li, W. S. Lee, and Philip Y. Building Text Classifiers Using Positive and Unlabeled Examples [C]. Proceedings of the Third IEEE International Conference on Data Mining (ICDM-03), Melbourne, Florida, November 2003, 19-22.
[53] H. Yu, J. Han, and Chang K.C.-C. PEBL: Positive Example Based Learning for Web Page Classification Using SVM [C]. Proc. Eighth Int'l Conf. Knowledge Discovery and Data Mining (KDD'02), 2002, 239-248.
[54] X.L. Li and B. Liu. Learning to Classify Text using Positive and Unlabeled Data [C]. Proceedings of Eighteenth International Joint Conference on Artificial Intelligence (IJCAI-03), Aug 9-15, 2003.
[55] G.P.C. Fung and H.J. Lu. Text Classification without Negative Examples Revisit [J]. IEEE Transactions on Knowledge and Data Engineering, 18(1): 6-20, 2006.
[56] F. Denis, R. Gilleron, M. Tommasi. Text Classification from Positive and Unlabeled Examples [C]. The 9th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2002.
[57] F. Denis, R. Gilleron, A. Laurent, M. Tommasi. Text Classification and Co-Training from Positive and Unlabeled Examples [C]. Proceedings of the ICML 2003 Workshop: The Continuum from Labeled to Unlabeled Data, 80-87, 2003.
[58] W.S. Lee, B. Liu. Learning with Positive and Unlabeled Examples using Weighted Logistic Regression [C]. Proceedings of the Twentieth International Conference on Machine Learning (ICML-2003), August 21-24, 2003.
[59] D. Zhang, W.S. Lee. A Simple Probabilistic Approach to Learning from Positive and Unlabeled Examples [C]. Proceedings of the 5th Annual UK Workshop on Computational Intelligence (UKCI), London, UK, September 2005.
[60] Dell Zhang, Wee Sun Lee. Learning Classifiers without Negative Examples: A Reduction Approach [C]. Proceedings of the 3rd IEEE International Conference on Digital Information Management (ICDIM), London, UK, Nov 2008.
[61] J., Hroza and J. ?i?ka, B. Pouliquen, C. Ignat and R. Steinberger. Mining Relevant Text Documents Using Ranking-Based k-NN Algorithms Trained by Only Positive Examples [C]. Proceedings of the Fourth Czech-Slovak Conference Knowledge-2005, February 9-11, 2005, StaráLesná, Slovak Republic, 29-40.
[62] J., Hroza, J. ?i?ka, B. Pouliquen, C. Ignat and R. Steinberger. The Selection of Electronic Text Documents Supported by Only Positive Examples [C]. JADT 2006, Besancon, France, 1001-1010.
[63] Aas K., Eikvil L. Text Categorization: A survey [R]. Technical Report, NR 941, Oslo: Norwegian Computing Center, 1999.
[64] Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze. Introduction to Information Retrieval [M]. Cambridge University Press, 2008.
[65] Sebastiani F. Machine Learning in Automated Text Categorization [J]. ACM Computing Surveys, 2002, 34(1): 1?47.
[66] Baeza-Yates, R. and B. Ribeiro-Neto. eds. Modern Information Retrieval [J]. ACM Press, 1999.
[67] G. Salton and M. J. McGill. An Introduction to Modern Information Retrieval [J]. McGraw-Hill, 1983.
[68] Popovic, M. and Willett, P. The Effectiveness of Stemming for Natural Language Access to Slovene Textual Data [J]. Journal of the American Society for Information Science, 43(5), 384-390, 1992.
[69] M. F. Porter. An Algorithm for Suffix Stripping [J]. Program, 14(3) pp 130?137, 1980.
[70] Manning C.D., Hinrich Schutze著,苑春法等译.统计自然语言处理基础[M].北京:电子工业出版社, 2005年1月.
[71] G. Salton, M. E. Lesk. Computer Evaluation of Indexing and Text Processing [J]. Journal of the ACM (JACM), ACM Press, 15(1) : 8-36, January 1968.
[72] Salton G. and Buckley C. Term-Weighting Approaches in Automatic Text Retrieval [J]. Information Processing and Management, 1988, 24(5): 513-523.
[73] Buckley C, Salton G et al. Automatic Query Expansion using SMART: TREC 3 [C]. In: Proc. 3rd Text Retrieval Conference, NIST, 1994.
[74] Dumais S. T. Improving the Retrieval Information from External Sources, Behaviour Research Methods [J]. Instruments and Computers, 1991, 23(2): 229-236.
[75] Y. Yang and J. O. Pedersen. A Comparative Study on Feature Selection in Text Categorization [C]. Proceedings of the Fourteenth International Conference on Machine Learning (ICML'97), 1997.
[76] Rocchio J. Relevance Feedback in Information Retrieval [C]. In The Smart Retrieval System: Experiments in Automatic Document Processing, G. Salton, Ed. Prentice-Hall, Englewood Cliffs, NJ, 1971, 313-323.
[77] Belur V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques [M]. McGraw Hill Computer Science Series, IEEE Computer Society Press, Las Alamitos, California, 1996.
[78] Y. Yang and X. Liu. A Re-Examination of Text Categorization Methods [C]. Proceedings of ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR'99), 1999.
[79] Joachims, Thorsten. A Probabilistic Analysis of the Rocchio Algorithm with TFIDF for Text Categorization [C]. Proceedings of International Conference on Machine Learning (ICML), 1997.
[80] A. McCallum and K. Nigam. A Comparison of Event Models for Naive Bayes Text Classification [C]. In AAAI-98 Workshop on Learning for Text Categorization, 1998.
[81] Vapnik V. Nature of Statistical Learning Theory (2nd edition) [M]. New York, Springer Press, 2000.
[82] Joachims T. Text Categorization with Support Vector Machines: Learning with ManyRelevant Features [C]. Proceedings of 10th European Conference on Machine Learning(ECML-98), Chemnitz, DE, 1998, 137-142.
[83]史忠植.知识发现[M].北京:清华大学出版社, 2002年1月.
[84] Fran?ois Denis, Rémi Gilleronb, and Fabien Letouzeyb. Learning from Positive and Unlabeled Examples [J]. Theoretical Computer Science, Volume 348, Issue 1, 2 December 2005, Pages 70-83.
[85] Larry M. Manevitz, Malik Yousef. One-Class SVMs for Document Classification [J]. Journal of Machine Learning Research, 2001(12), 139-154.
[86] Hwanjo Yu, Jiawei Han, Kevin Chen-Chuan Chang. PEBL: Web Page Classification without Negative Examples [J]. IEEE Transactions on Knowledge and Data Engineering, 16(1): 70-81, January 2004.
[87] Hwanjo Yu. Single-Class Classification with Mapping Convergence[J]. Machine Learning, 61(1-3): 49-69, November 2005.
[88] Rakesh Agrawal, Roberto Bayardo, Ramakrishnan Srikant, Athena. Mining-based Interactive Management of Text Databases [C]. International Conference on Extending Database Technology, 2000, 365-379.
[89] Buckley, C., Salton G., and Allan J. The effect of adding relevance information in a relevance feedback environment [C]. SIGIR94, 1994.
[90] A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum Likelihood from Incomplete Data via the EM Algorithm [J]. J. Royal Stat. Soc., 39: 1-38, 1977.
[91] Hailong Yu, Wanli Zuo, Tao Peng. A New PU Learning Algorithm for Text Classification [C]. MICAI 2005: 824-832.
[92] Tao Peng, Wanli Zuo, Fengling He. SVM based Adaptive Learning Method for Text Classification from Positive and Unlabeled Documents [J]. Knowl. Inf. Syst. 16(3): 281-301, 2008.
[93] Hui Wang, and Wanli Zuo. Extracting Initial and Reliable Negative Documents to Enhance Classification Performance [C]. KDLL 2006, 104-111.
[94] Gabriel Pui Cheong Fung, Jeffrey Xu Yu, Hongjun Lu, Philip S. Yu. Text Classification without Labeled Negative Documents [C]. ICDE 2005, 594-605.
[95] Shuang Yu, Chunping Li. PE-PUC: A Graph based PU-Learning Approach for Text Classification [C]. 5th International Conference on Machine Learning and Data Mining in Pattern Recognition, Leipzig, Germany, Lecture Notes in Computer Science 4571, 574 -584, 2007.
[96] Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal. Semi-Supervised Learning on Directed Graphs [C]. Advances in Neural Information Processing Systems, 2005, 1633--1640.
[97] T. Joachims. A Support Vector Method for Multivariate Performance Measures [C]. Proceedings of the International Conference on Machine Learning (ICML), 2005.
[98] Xiaoli Li, and Bing Liu. Dealing with Different Distributions in Learning from Positive and Unlabeled Web Data [C]. WWW-2004.
[99] Xiaoli Li, and Bing Liu. Learning from Positive and Unlabeled Examples with Different DataDistributions [C]. European Conference on Machine Learning (ECML-05), Porto, Portugal, LNCS 3720, Springer, October 3-7, 2005, 218-229.
[100] Bing Liu, XiaoLi Li, W.S. Lee and P.S. Yu. Text Classification by Labeling Words [C]. Proceedings of Nineteenth National Conference on Artificial Intelligence (AAAI-2004), San Jose, California, USA. AAAI Press/MIT Press, July 25-29, 2004, 425-430.
[101] Gao Cong, Wee Sun Lee, Haoran Wu, Bing Liu. Semi-Supervised Text Classification Using Partitioned EM [C]. DASFAA 2004: 482-493.
[102] Borja Calvo, Paseo Manuel de Lardizabal, Pedro Larra?aga, Paseo Manuel de Lardizabal, JoséA. Lozano, Paseo Manuel de Lardizabal. Learning Bayesian Classifiers from Positive and Unlabeled Examples [J]. Pattern Recognition Letters archive, Volume 28, Issue 16, 2375-2384, 2007.
[103] Tao Peng, Fengling He, Wanli Zuo, Changli Zhang. Adaptive Topical Web Crawling for Domain-Specific Resource Discovery Guided by Link-Context [C]. MICAI 2006: 963-973.
[104] Xiaoli Li, Bing Liu, and S.-K. Ng. Learning to Identify Unexpected Instances in the Test Set [C]. Proceedings of Twenty International Joint Conference on Artificial Intelligence (IJCAI-07), Hyderabad, India, January 6-12, 2007, 2802-2807.
[105] Zhi-Hua Zhou. Learning with Unlabeled Data and Its Application to Image Retrieval [J]. Lecture Notes in Computer Science, Volume 4099/2006, Springer Berlin / Heidelberg, 2006, 5-10.
[106] Karl-Michael Schneider. Learning to Filter Junk E-Mail from Positive and Unlabeled Examples [J]. Lecture Notes in Computer Science, Springer Berlin / Heidelberg, Volume 3248/2005, 426-435.
[107] Wang C, Ding C, Meraz RF, Holbrook SR. PSoL: A Positive Sample Only Learning Algorithm for Finding Non-Coding RNA Genes [J]. Bioinformatics, 22(21): 2590-6, 2006.
[108] Charles Elkan, Keith Noto. Learning Classifiers from Only Positive and Unlabeled Data [C]. Proceeding of the 14th ACM SIGKDD, Las Vegas, Nevada, USA, 213-220, 2008.
[109]苏金树,张博锋,徐昕.基于机器学习的文本分类技术研究进展[J].软件学报, September 2006, 17(9): 1848?1859.
[110] Reuters-21578语料[EB/OL]. http://www.research.att.com/~lewis.
[111] Bow: A Toolkit for Statistical Language Modeling, Text Retrieval, Classification and Clustering [EB/OL]. http://www.cs.cmu.edu/~mccallum/bow/.
[112]周志华.半监督学习专刊前言[J].软件学报, 2008, 19(11): 2789?2790.
[113] D. J. Miller, H. S. Uyar. A Mixture of Experts Classifier with Learning based on both Labeled and Unlabelled Data [C]. In: M. Mozer, M. I. Jordan, T. Petsche, eds. Advances in Neural Information Processing Systems 9, Cambridge, MA: MIT Press, 1997, 571-577.
[114] T. Zhang, F. J. Oles. A Probability Analysis on the Value of Unlabeled Data for Classification Problems [C]. In: Proceedings of the 17th International Conference on Machine Learning (ICML’00), San Francisco, CA, 2000, 1191-1198.
[115] X. Zhu. Semi-Supervised Learning Literature Survey [R]. Technical Report 1530, Department of Computer Sciences, University of Wisconsin at Madison, Madison, WI, July19, 2008.
[116] Thorsten Joachims. Transductive Inference for Text Classification using Support Vector Machines[J]. Morgan Kaufmann, 1999, 200-209.
[117] Brendan J. Frey, Delbert Dueck. Clustering by Passing Messages Between Data Points, [J]. Science, 16 February 2007, Vol. 315, no. 5814, 2007, 972– 976.
[118] Kyriakopoulou, A. Text Classification Aided by Clustering: a Literature Review [M]. I-Tech Education and Publishing KG, Vienna, Austria (2008)
[119] Kyriakopoulou, A., Kalamboukis, T. Using Clustering to Enhance Text Classification [C]. In SIGIR 2007, 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 805–806, 2007.
[120] Kyriakopoulou, A., Kalamboukis, T. Text Classification using Clustering [C]. In: ECML/PKDD Discovery Challenge Workshop, 2006.
[121] Kyriakopoulou, A. Using Clustering and Co-Training to Boost Classification Performance [C]. In: ICTAI (2). IEEE Computer Society, 2007, 325–330.
[122] Tishby, N. Z., Pereira, F., Bialek, W. The Information Bottleneck Method [C]. In Proceedings of the 37th Allerton Conference on Communication, Control and Computing, 1999.
[123] Pereira F., Tishby N., Lee L. Distributional Clustering of English Words [C]. Proceedings of the 31st Annual Meeting of the Association for Computational Linguistics, 1993, 183-190.
[124] Baker L. D., McCallum A. K., (1998) Distributional Clustering of Words for Text Classification [C]. Proceedings of SIGIR’98, 21st ACM International Conference on Research and Development in Information Retrieval, pages 96–103, Melbourne, AU. ACM Press, New York, US.
[125] Dhillon I., Mallela S., Kumar R. A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification [J]. Journal of Machine Learning Research 3, 2003, 1265-1287.
[126] Slonim, N., Tishby, N. Document Clustering using Word Clusters via the Information Bottleneck Method[C]. In Proceedings of the ACM SIGIR, 2000.
[127] Yaniv R. E., Souroujon O. Iterative Double Clustering for Unsupervised and Semi-Supervised Learning [C]. In proceedings of the 12th European Conference on Machine Learning, ECML, 2001.
[128] J. Shi, J. Malik. Normalized Cuts and Image Segmentation [C]. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR'97), 731-737, 1997.
[129] Dhillon I. Co-Clustering Document and Word using Bipartite Spectral Graph Partitioning [C]. In: Proceedings of the ACM SIGKDD’01, 2001, 269-274.
[130] Dhillon I., Mallela S., Modha, S. Information Theoretic Co-Clustering [C] In Proceedings of the ACM SIGKDD Conference, 2003.
[131] Fung, G. and Mangasarian, O.L. Semi-Supervised Support Vector Machines for Unlabeled Data Classification [J]. Optim. Methods Software, 2001, 15(1): 29-44.
[132] Li, M., Cheng, Y., Zhao, H. Unlabeled Data Classification via Support Vector Machine andk-means Clustering[C]. In Proceedings of the International Conference on Computer Graphics, Imaging and Visualization, CGIV'04, 2004, 183-186.
[133] Hua-Jun Zeng, Xuan-Hui Wang, Zheng Chen, Hongjun Lu, Wei-Ying Ma. CBC: Clustering based Text Classification Requiring Minimal Labeled Data [C]. Third IEEE International Conference on Data Mining, 2003, Nov. 2003, 443– 450.
[134] Chapelle, O., Weston, J., Scholkopf, B. Cluster Kernels for Semi-Supervised Learning [C]. In NIPS, volume 15, 2002.
[135] Raskutti, B., Ferrá, H., Kowalczyk, A. Using Unlabelled Data for Text Classification through Addition of Cluster Parameters [C]. Proceedings of the Nineteenth International Conference on Machine Learning, 2000, 514–521.
[136] Raskutti, B., Ferrá, H., Kowalczyk, A. Combining Clustering and Co-Training to Enhance Text Classification using Unlabelled Data [C]. Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining, 2002.
[137] Yu, H., Yang, J., Han, J. Classifying Large Data Sets using SVMs with Hierarchical Clusters [C]. in Proceedings of the 9th ACM SIGKDD 2003, Washington, DC, USA, 2003.
[138] Boley, D., Cao, D. Training Support Vector Machine using Adaptive Clustering [C]. Proceeding of 2004 SIAM International Conference on Data Mining, 2004.
[139] Asharaf, S., Murty, M. N., Shevade, S. K. Cluster based Training for Scaling Nonlinear Support Vector Machines [C]. Proceedings of the International Conference on Computing: Theory and Applications (ICCTA'07), 2007.
[140] Li, B., Chi, M., Fan, J., Xue, X. Support Cluster Machine [C]. Proceedings of the 24th International Conference on Machine Learning, Corvallis, OR, 2007.
[141] Cervantes, J., Li, X., Yu, W., Li, K. Support Vector Machine Classification for Large Data Sets via Minimum Enclosing Ball Clustering [J]. Neurocomputing, 71(4-6): 611-619, 2008.
[142] Sun, S., Tseng, C. L., Chen, Y. H., Chuang, S. C., Fu, H. C. Cluster-based Support Vector Machines in Text-Independent Speaker Identification [C]. In Proceedings of the International Joint Conference on Neural Network, 2004.
[143] Wang, J., Wu, X., Zhang, C. Support Vector Machines based on k-Means Clustering for Real-Time Business Intelligence Systems [C]. Int. J. Business Intelligence and Data, Mining, 1(1): 54–64, 2005.
[144] He, J., Zhong, W., Harrison, R., Tai, P. C., Pan, Y. Clustering Support Vector Machines and Its Application to Local Protein Tertiary Structure Prediction [C]. ICCS 2006, part II, LNCS 3992, 2006, 710-717.
[145] CLUTO - Software for Clustering High-Dimensional Datasets [EB/OL]. http://glaros.dtc.umn.edu/ gkhome/ cluto/cluto/download.
[146] Thorsten Joachims. SVMlight Support Vector Machine [EB/OL]. http://svmlight.joachims.org/.
[147] W. Tang, H. Xiong, S. Zhong, and J. Wu. Enhancing Semi-Supervised Clustering: A Feature Projection Perspective [C]. Proceedings of the 13th ACM SIGKDD, 2007, 707-716.
[148] Demiriz A, Bennett KP, Embrechts MJ. Semi-Supervised Clustering using GeneticAlgorithms [C]. In: Dagli CH, ed. Proc. of the Intelligent Engineering Systems Through Artificial Neural Networks 9 (ANNIE’99). New York: ASME Press, 1999, 809?814.
[149] Wagstaff K, Cardie C, Rogers S, Schroedl S. Constrained K-means Clustering with Background Knowledge [C]. In: Carla EB, Andrea PD, eds. Proc. of the 18th Int’l Conf. on Machine Learning (ICML 2001). San Fransisco: Morgan Kaufmann Publishers, 2001, 577?584.
[150] Basu S, Bilenko M, Mooney RJ. A Probabilistic Framework for Semi-Supervised Clustering [C]. In: Won K, Ron K, Johannes G William D, eds. Proc. of the 10th ACM SIGKDD Int’l Conf. on Knowledge Discovery and Data Mining (KDD 2004). Seattle: ACM Press, 2004, 59?68.
[151] Basu S, Banerjee A, Mooney RJ. Semi-Supervised Clustering by Seeding [C]. In: Claude S, Achim GH, eds. Proc. of the 19th Int’l Conf. on Machine Learning (ICML 2002), San Fransisco: Morgan Kaufmann Publishers, 2002, 19?26.
[152] Basu S. Semi-Supervised clustering: Probabilistic Models Algorithms and Experiments [D]. Austin: University of Texas at Austin, 2005.
[153] Xing EP, Ng AY, Jordan MI, Russell S. Distance Metric Learning with Application to Clustering with Side-information [C]. In: Becher S, Thrun S, Obermayer K, eds. Proc. of the 16th Annual Conf. on Neural Information Processing System, Cambridge: MIT Press, 2003, 505?512.
[154] Bar-Hillel A, Hertz T, Shental N, Weinshall D. Learning Distance Functions using Equivalence Relations [C]. In: Fawcett T, Mishra N, eds. Proc. of the 20th Int’l Conf. on Machine Learning, Washington: Morgan Kaufmann Publishers, 2003, 11?18.
[155] Yeung DY, Chang H. Extending the Relevant Component Analysis Algorithm for Metric Learning using both Positive and Negative Equivalence Constraints [J]. Pattern Recognition, 2006, 39(5):1007?1010.
[156] Schultz M, Joachims T. Learning a Distance Metric from Relative Comparisons [C]. In: Thrun S, Saul LK, Sch?lkopf B, eds. Proc. of the 17th Annual Conf. on Neural Information Processing System, Cambridge: MIT Press, 2004, 41?48.
[157] De la Torre F, Kanade T. Discriminative Cluster Analysis [C]. In: William WC, Andrew M, eds. Proc. of the 19th Int’l Conf. on Machine Learning, New York: ACM Press, 2006, 241?248.
[158] Ding CH, Li T. Adaptive Dimension Reduction using Discriminant Analysis and K-means Clustering [C]. In: Ghahramani Z, ed. Proc. of the 19th Int’l Conf. on Machine Learning, New York: ACM Press, 2007, 521?528.
[159] Ye JP, Zhao Z, Liu H. Adaptive Distance Metric Learning for Clustering [C]. In: Bishop CM, Frey B, eds. Proc. of IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, Madison: IEEE Computer Society Press, 2007, 1?7.
[160]尹学松,胡恩良,陈松灿.基于成对约束的判别型半监督聚类分析[C].软件学报, 2008, 19(11): 2791?2802.
[161] Bilenko M, Basu S, Mooney RJ. Integrating Constraints and Metric Learning inSemi-Supervised Clustering [C]. In: Brodley CE, ed. Proc. of the 21st Int’l Conf. on Machine Learning, New York: ACM Press, 2004, 81?88.
[162] Basu S, Banerjee A, Mooney RJ. A Probabilistic Framework for Semi-Supervised Clustering [C]. In: Boulicaut JF, Esposito F, Giannotti F, Pedreschi D, eds. Proc. of the 10th ACM SIGKDD Int’l Conf. on Knowledge Discovery and Data Mining, New York: ACM Press, 2004, 59?68.
[163] Brian Kulis, Sugato Basu, Inderjit Dhillon, and Raymond Mooney. Semi-Supervised Graph Clustering: A Kernel Approach [J]. Machine Learning Journal, 74(1): 1-22, 2009.
[164] Foster Provost, Pedro Domingos. Tree Induction for Probability-based Ranking, [J]. Machine Learning, 2003, 199-215.
[165] Maytal Saar-tsechansky, Foster Provost. Active Sampling for Class Probability Estimation and Ranking [J]. Machine Learning, 2004, 153-178.
[166] Chris Burges, Tal Shaked, Erin Renshaw, Ari Lazier, Matt Deeds, Nicole Hamilton, Greg Hullender. Learning to Rank using Gradient Descent [C]. in Proceedings of the 22nd International Conference on Machine Learning, Bonn, Germany, 2005.
[167] Pinar Donmez, Jaime G. Carbonell. Optimizing Estimated Loss Reduction for Active Sampling in Rank Learning: DiffLoss [C]. ICML 2008, Helsinki, Finland, June 7, 2008.
[168] J. Xu and H. Li. Adarank: A Boosting Algorithm for Information Retrieval [C]. In SIGIR 2007, 391-398, 2007.
[169] Hroza, J., ?i?ka, J. Selecting Interesting Articles Using Their Similarity Based Only on Positive Examples [C]. Proceedings of the Sixth International Conference on Intelligent Text Processing and Computational Linguistics CICLing-2005, February 13-19, 2005, Mexico City, Mexico, Springer-Verlag, 2005, LNCS/LNAI 3406, 608- 611.
[170] Kevin Duh, Katrin Kirchhoff. Learning to Rank with Partially-Labeled Data [C]. Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval, Singapore, 2008, 251-258.
[171] Y. Yang. An Evaluation of Statistical Approaches to Text Categorization, Information Retrieval archive [M]. Kluwer Academic Publishers, Hingham, MA, USA, 1(1-2): 69-90, 1999.
[172] Kamal Nigam, Rayid Ghani. Analyzing the Effectiveness and Applicability of Co-Training [C]. In Proc. of Ninth International Conference on Information and Knowledge (CIKM-2000).
[173] Pierce, David and Cardie, Claire. Limitations of Co-Training for Natural Language Learning from Large Datasets [C]. Proceedings of the 2001 Conference on Empirical Methods in Natural Language Processing, 2001, 1-9.
[174] S. Dasgupta, M. Littman, D. McAllester. PAC Generalization Bounds for Co-Training [C]. In: T. G. Dietterich, S. Becker, Z. Ghahramani, eds. Advances in Neural Information Processing Systems 14, Cambridge, MA: MIT Press, 2002, 375-382.
[175] M.-F. Balcan, A. Blum, K. Yang. Co-Training and Expansion: Towards Bridging Theory and Practice [C]. In: L. K. Saul, Y. Weiss, L. Bottou, eds. Advances in Neural Information Processing Systems 17, Cambridge, MA: MIT Press, 2005, 89-96.
[176] W. Wang, Z.-H. Zhou. Analyzing Co-Training Style Algorithms [C]. In: Proceedings of the 18th European Conference on Machine Learning (ECML’07), Warsaw, Poland, 2007.
[177] Z.-H. Zhou, M. Li. Semi-Supervised Learning with Co-Training [C]. In: Proceedings of the 19th International Joint Conference on Artificial Intelligence (IJCAI’05), Edinburgh, Scotland, 2005, 908-913.
[178] Z.-H. Zhou, M. Li. Semi-Supervised Learning with Co-Training Style Algorithm [J]. IEEE Transactions on Knowledge and Data Engineering, 19(11), 2007.
[179] U. Brefeld, T. G?rtner, T. Scheffer, S. Wrobel. Efficient Co-Regularised Least Squares Regression [C]. In: Proceedings of the 23rd International Conference on Machine Learning (ICML’06), Pittsburgh, PA, 2006, 137-144.
[180] D. Yarowsky. Unsupervised Word Sense Disambiguation Rivaling Supervised Methods [C]. In: Proceedings of the 33rd Annual Meeting of the Association for Computational Linguistics (ACL’95), Cambridge, MA, 1995, 189-196.
[181] S., Kiritchenko, S. Matwin. Email Classification with Co-training [R]. (Technical Report), University of Ottawa, 2002.
[182] Jason Chan, Irena Koprinska, Josiah Poon. Co-training with a Single Natural Feature Set Applied to Email Classification [C]. Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence, 2004, 586-589.
[183]张博锋,苏金树,徐昕.文本分类中用于协同训练的特征集分割[J/OL].中国科技论文在线, http://www.edu.cn, 2008-04-01.
[184] Ion Muslea, Steven Minton, Craig A. Knoblock. Adaptive View Validation: A First Step Towards Automatic View Detection [C]. Proceedings of ICML 2002, 443-450.
[185] M. Li, Z.-H. Zhou. SETRED: Self-Training with Editing [C]. Proceedings of 9th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD’05), Hanoi, Vietnam, LNAI 3518, 2005, 611-621.
[186] Ion Muslea, Steve Minton, Craig Knoblock. Selective Sampling with Redundant Views, [C]. Proceedings of the Fifteenth National Conference on Artificial Intelligence, AAAI-2000, 621-626.
[187] Ulf Brefeld, Tobias Scheffer. Co-EM Support Vector Learning [C]. Proceedings of the twenty-first international conference on Machine learning, Banff, Alberta, Canada, 2004.
[188] S. Goldman, Y. Zhou. Enhancing Supervised Learning with Unlabeled Data [C]. In: Proceedings of the 17th International Conference on Machine Learning (ICML’00), San Francisco, CA, 2000, 327-334.
[189] D. Angluin, P. Laird. Learning from Noisy Examples [J]. Machine Learning, 1988, 2(4): 343-370.
[190] Y. Zhou, S. Goldman. Democratic Co-learning [C]. In: Proceedings of the 16th IEEE International Conference on Tools with Artificial Intelligence (ICTAI’04), Boca Raton, FL, 2004, 594–602.
[191] D. Mavroeidis, K. Chaidos, S. Pirillos, D. Christopoulos, M. Vazirgiannis. Using Tri-Training and Support Vector Machines for Addressing the ECML/PKDD 2006 DiscoveryChallenge [C]. In: Proceedings of the ECML-PKDD Discovery Challenge Workshop, Berlin, Germany, 2006, 39-47.
[192] N. T. Tri, N. M. Le, and Akira S. Using Semi-Supervised Learning for Question Classification [C]. Lecture Notes in Computer Science, Volume 4285, 2006, 31-41.
[193] W. L. Chen, Y. J. Zhang, and Isahara H. Chinese Chunking with Tri-Training Learning [C]. 21st International Conference on the Computer Processing of Oriental Languages (ICCPOL2006), 2006, 466-473.
[194]邓超,郭茂祖.基于自适应数据剪辑策略的Tri-training算法[J].计算机学报, 2007年8月, 30(8): 1213-1226.
[195]邓超,郭茂祖.基于Tri-training和数据剪辑的半监督聚类算法[J].软件学报, 2008, 19(3): 663-673.
[196] M. Li, Z.-H. Zhou. Improve Computer-aided Diagnosis with Machine Learning Techniques using Undiagnosed Samples [J]. IEEE Transactions on Systems, Man and Cybernetics - Part A: Systems and Humans, 2007, 37(6): 1088-1098.
2 https://sites.google.com/site/googol/Home/google-news/2008-%E5%B9%B4-7-%E6%9C%88
1 http://www.cnnic.net.cn/uploadfiles/doc/2009/1/13/92209.doc
[1] Soumen Chakrabarti, M. van den Berg and B. Dom.Focused Crawling: A New Approach to Topic-Specific Web Resource Discovery [J]. Computer Networks, 1999, 31(11-16): 1623–1640.
[2] B. Liu. Web Data Mining Exploring Hyperlinks, Contents and Usage Data [M], Springer, December, 2006.
[3]李晓明,闫宏飞,王继民.搜索引擎-原理、技术与系统[M].北京:科学出版社,2004.
[4] S. Brin and L. Page.The Anatomy of a Large-scale Hypertextual Web Search Engine [J]. Computer Networks, 1998, 30(1-7): 107–117.
[5]王珏,周志华,周傲英.机器学习及其应用[C].北京:清华大学出版社,2006.
[6] Soumen Chakrabarti . Mining the Web Discovering Knowledge from Hypertext Data [M].Morgan-Kaufmann Publishers, 2002.
[7] Fabrizio Sebastiani. A Tutorial on Automated Text Categorization [C], In Analia Amandi and Alejandro Zunino (eds.), Proceedings of ASAI-99, 1st Argentinian Symposium on Artificial Intelligence, Buenos Aires, AR, 1999, 7-35.
[8] J. M. Kleinberg. Authoritative Sources in a Hyperlinked Environment [J], J. ACM, 1999, 46(5): 604–632.
[9] B. D. Davison. Topical Locality in the Web [C], SIGIR2000, 272–279, 2000.
[10] S. Chakrabarti, K. Punera and M. Subramanyam. Accelerated Focused Crawling through Online Relevance Feedback [C], WWW2002, 148–159, 2000.
[11] P. D. Bra, G. Houben, Y. Kornatzky, and R. Post. Information Retrieval in Distributed Hypertexts [C], Procs. of the 4th RIAO Conference, New York, 481–491, 1994.
[12] P D Bra, et al. Searching for Arbitrary Information in the WWW: The Fish-Search for Mosac [C], WWW Conference, 1994.
[13] M. Hersovici, M. Jacovi, Y. S. Maarek, D. Pellegb, M. Shtalhaima, and S. Ura. The Shark-Search Algorithm, An Application: Tailored Web Site Mapping [C]. In WWW7. 1998.
[14] J. Cho, H. Efficient Crawling Through URL Ordering [C], Garcia-Molina, L. Page. In Proceedings of the 7th International WWW Conference, Brisbane, Australia, April 1998.
[15] L. Page, S. Brin, R. Motwani, T. Winograd. The PageRank Citation Ranking: Bringing Order to the Web [R], Stanford Digital Library Technologies Project, 1998.
[16] Menczer F,Pant G,Ruiz M,Srinivasan P. Evaluating Topic-Driven Web Crawlers [C]. In: 24th Annual Intl. ACM SIGIR Conf. on Research and Development in Information Retrieval, 2001, New York, 241-249.
[17] S. Chakrabarti, David A. Gibson, and Kevin S. Mccurley. Surfing the Web Backwards [C], In Proceedings of the 8th International WWW Conference. Toronto, Canada, 1679-1693 May 1999.
[18] A. McCallum, K. Nigam, J. Rennie, and K. Seymore. Building Domain-Specific Search Engines with Machine Learning Technique [C], In Procs. of AAAI Spring Symposium on Intelligents Engine in Cyberspace, 1999.
[19] Rennie J.,McCallum A.Using Reinforcement Learning to Spider the Web Efficiently [C], In:Proceedings of ICML-99, 16th International Conference on Machine Learning, Bled, Slovenia, 1999, 335-343.
[20] M. Diligenti, F. M. Coetzee, S. Lawrence, C. L. Giles, and M. Gori, Focused Crawling using Context Graphs [C], In Procs. of the 26th VLDB Conference, Cairo, Egypt, 2000.
[21] F Menczer, G Gant, P Srinivasan. Topic-Driven Crawlers: Machine Learning Issues [C], ACM TOIT, 2002.
[22] Chiasen Chung, Charles L. A. Clarke. Topic-Oriented Collaborative Crawling [C], CIKM’02, November McLean, Virginia, USA, 2002.
[23] Jan Fiedler, Joachim Hammer. Using the Web Efficiently: Mobile Crawlers[C], Proc of 7th AoM/IaoM Intl Confrence on Computer Science, San Diego, CA, 1999.
[24] M. Ehrig, A. Maedche. Ontology-Focused Crawling of Web Documents [C], Proceedings of the 2003 ACM symposium on Applied computing, 2003.
[25] Silva, I., B. Ribeiro-Neto, P. Calado, N. Ziviani, and E. Moura. Link-based and Content-based Evidential Information in A Belief Network Model [C], Proceedings of the 23rd International ACM SIGIR Conference on Research and Development in Information Retrieval, 96-103, 2000.
[26] Amento, B., L. Terveen, and W. Hill. Does“Authority”Mean Quality? Predicting Expert Quality Ratings of Web Documents[C]. Proc. 23rd ACM SIGIR Conf. on Research and Development in Information Retrieval, 296-303, 2000.
[27] Pant, G., P. Srinivasan, and F. Menczer. Exploration versus Exploitation in Topic Driven Crawlers [C]. In: Proc. WWW-02 Workshop on Web Dynamics, 2002.
[28] Spink, A., D. Wolfram, B. Jansen, and T. Saracevic. Searching the Web: The Public and Their Queries [J]. Journal of the American Society for Information Science, 52(3):226–234, 2001.
[29]黄菁萱,吴立德.基于向量空间模型的文档分类系统[J].模式识别与人工智能, 11(2), 1998
[30] Shih LK, Karger DR. Using URLs and Table Layout for Web Classification Tasks [C]. Proc. of the 13th Int’l Conf. on the World Wide Web (WWW-2004), 193-202, 2004.
[31] Tom M. Mitchell. Machine Learning [M]. McGraw-Hill Companies, Inc, 1997.
[32] O. Chapelle and B. Schokopf and A. Zien. Semi-Supervised Learning [M]. MIT Press, Cambridge, MA,2006.
[33] Jiawei Han, Micheline Kamber著,范明,孟小峰译.数据挖掘概念与技术[M].第二版,北京:机械工业出版社,2007年3月.
[34] Pang-Ning Tan, Michael Steinbach, Vipin Kumar著,范明,范宏建译.数据挖掘导论[M].北京:人民邮电出版社, 2006年5月.
[35]周志华,王珏.机器学习及其应用2007 [M].北京:清华大学出版社, 2007.
[36] B. Shahshahani, D. Landgrebe, The Effect of Unlabeled Samples in Reducing the Small Sample Size Problem and Mitigating the Hughes Phenomenon [J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(5): 1087-1095.
[37] R. P. Lippmann. Pattern Classification using Neural Networks [J]. IEEE Communications, 1989, 27(11): 47-64.
[38] D. J. Miller, H. S. Uyar. A Mixture of Experts Classifier with Learning based on Both Labeled and Unlabeled Data [C]. In: M. Mozer, M. I. Jordan, T. Petsche, eds, Advances in Neural Information Processing Systems 9, Cambridge, MA: MIT Press, 1997, 571-577.
[39] A. Mccallum, K. Nigam. Text Classification by Bootstrapping with Keywords, EM and Shrinkage [C]. Workshop On Unsupervised Learning In Natural Language Processing, 1999, 52-58.
[40] K. Nigam, A. K. McCallum, S. Thrun, T. Mitchell. Using EM to Classify Text from Labeled and Unlabeled Documents [J], Machine Learning, 1998, AAAI Press, 39-103.
[41] K. Nigam, A. K. McCallum, S. Thrun, T. Mitchell. Text Classification from Labeled and Unlabeled Documents using EM [J]. Machine Learning, 2000, 39(2-3): 103-134.
[42] A. Blum, S. Chawla. Learning from Labeled and Unlabeled Data using Graph Mincuts [C]. Proceedings of the 18th International Conference on Machine Learning (ICML’01), San Francisco, CA, 2001, 19-26.
[43] X. Zhu, Z. Ghahramani, J. Lafferty. Semi-Supervised Learning using Gaussian Fields and Harmonic Functions [C] In: Proceedings of the 20th International Conference on Machine Learning (ICML’03), Washington, DC, 2003, 912-919.
[44] M. Belkin, P. Niyogi. Semi-Supervised Learning on Riemannian Manifolds [J]. Machine Learning, 2004, 56(1-3): 209-239.
[45] D. Zhou, O. Bousquet, T. N. Lal, J. Weston, B. Sch?lkopf. Learning with Local and Global Consistency [C]. In: S. Thrun, L. Saul, B. Sch?lkopf, eds. Advances in Neural Information Processing Systems 16, Cambridge, MA: MIT Press, 2004, 321-328.
[46] M. Belkin, P. Niyogi, V. Sindwani. On Manifold Regularization [C]. In: Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics (AISTATS’05), Savannah Hotel, Barbados, 2005, 17-24.
[47] A. Blum, T. Mitchell. Combining Labeled and Unlabeled Data with Co-Training [C]. In: Proceedings of the 11th Annual Conference on Computational Learning Theory (COLT’98), Wisconsin, MI, 1998, 92-100.
[48] Z.-H. Zhou and M. Li. Tri-Training: Exploiting Unlabeled Data using Three Classifiers [J]. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(11): 1529–1541.
[49] F. Denis. PAC Learning from Positive Statistical Queries [C]. The 9th International Workshop on Algorithmic Learning Theory (ALT'98), LNCS, Springer, Heidelberg, 1998, 112-126.
[50] B. Liu, W.S. Lee, P.S. Yu and X.L. Li. Partially Supervised Classification of Text Documents [C]. Proceedings of the Nineteenth International Conference on Machine Learning (ICML-2002), 8-12, July 2002.
[51] Bangzuo Zhang, Wanli Zuo. Learning from Positive and Unlabeled Examples: A Survey [C].2008 International Symposiums on Information Processing and 2008 International Pacific Workshop on Web Mining and Web-based Application,Moscow, Russia, May 23-25, 2008, 640-644.
[52] B. Liu, Y. Dai, X.L. Li, W. S. Lee, and Philip Y. Building Text Classifiers Using Positive and Unlabeled Examples [C]. Proceedings of the Third IEEE International Conference on Data Mining (ICDM-03), Melbourne, Florida, November 2003, 19-22.
[53] H. Yu, J. Han, and Chang K.C.-C. PEBL: Positive Example Based Learning for Web Page Classification Using SVM [C]. Proc. Eighth Int'l Conf. Knowledge Discovery and Data Mining (KDD'02), 2002, 239-248.
[54] X.L. Li and B. Liu. Learning to Classify Text using Positive and Unlabeled Data [C]. Proceedings of Eighteenth International Joint Conference on Artificial Intelligence (IJCAI-03), Aug 9-15, 2003.
[55] G.P.C. Fung and H.J. Lu. Text Classification without Negative Examples Revisit [J]. IEEE Transactions on Knowledge and Data Engineering, 18(1): 6-20, 2006.
[56] F. Denis, R. Gilleron, M. Tommasi. Text Classification from Positive and Unlabeled Examples [C]. The 9th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2002.
[57] F. Denis, R. Gilleron, A. Laurent, M. Tommasi. Text Classification and Co-Training from Positive and Unlabeled Examples [C]. Proceedings of the ICML 2003 Workshop: The Continuum from Labeled to Unlabeled Data, 80-87, 2003.
[58] W.S. Lee, B. Liu. Learning with Positive and Unlabeled Examples using Weighted Logistic Regression [C]. Proceedings of the Twentieth International Conference on Machine Learning (ICML-2003), August 21-24, 2003.
[59] D. Zhang, W.S. Lee. A Simple Probabilistic Approach to Learning from Positive and Unlabeled Examples [C]. Proceedings of the 5th Annual UK Workshop on Computational Intelligence (UKCI), London, UK, September 2005.
[60] Dell Zhang, Wee Sun Lee. Learning Classifiers without Negative Examples: A Reduction Approach [C]. Proceedings of the 3rd IEEE International Conference on Digital Information Management (ICDIM), London, UK, Nov 2008.
[61] J., Hroza and J. ?i?ka, B. Pouliquen, C. Ignat and R. Steinberger. Mining Relevant Text Documents Using Ranking-Based k-NN Algorithms Trained by Only Positive Examples [C]. Proceedings of the Fourth Czech-Slovak Conference Knowledge-2005, February 9-11, 2005, StaráLesná, Slovak Republic, 29-40.
[62] J., Hroza, J. ?i?ka, B. Pouliquen, C. Ignat and R. Steinberger. The Selection of Electronic Text Documents Supported by Only Positive Examples [C]. JADT 2006, Besancon, France, 1001-1010.
[63] Aas K., Eikvil L. Text Categorization: A survey [R]. Technical Report, NR 941, Oslo: Norwegian Computing Center, 1999.
[64] Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze. Introduction to Information Retrieval [M]. Cambridge University Press, 2008.
[65] Sebastiani F. Machine Learning in Automated Text Categorization [J]. ACM Computing Surveys, 2002, 34(1): 1?47.
[66] Baeza-Yates, R. and B. Ribeiro-Neto. eds. Modern Information Retrieval [J]. ACM Press, 1999.
[67] G. Salton and M. J. McGill. An Introduction to Modern Information Retrieval [J]. McGraw-Hill, 1983.
[68] Popovic, M. and Willett, P. The Effectiveness of Stemming for Natural Language Access to Slovene Textual Data [J]. Journal of the American Society for Information Science, 43(5), 384-390, 1992.
[69] M. F. Porter. An Algorithm for Suffix Stripping [J]. Program, 14(3) pp 130?137, 1980.
[70] Manning C.D., Hinrich Schutze著,苑春法等译.统计自然语言处理基础[M].北京:电子工业出版社, 2005年1月.
[71] G. Salton, M. E. Lesk. Computer Evaluation of Indexing and Text Processing [J]. Journal of the ACM (JACM), ACM Press, 15(1) : 8-36, January 1968.
[72] Salton G. and Buckley C. Term-Weighting Approaches in Automatic Text Retrieval [J]. Information Processing and Management, 1988, 24(5): 513-523.
[73] Buckley C, Salton G et al. Automatic Query Expansion using SMART: TREC 3 [C]. In: Proc. 3rd Text Retrieval Conference, NIST, 1994.
[74] Dumais S. T. Improving the Retrieval Information from External Sources, Behaviour Research Methods [J]. Instruments and Computers, 1991, 23(2): 229-236.
[75] Y. Yang and J. O. Pedersen. A Comparative Study on Feature Selection in Text Categorization [C]. Proceedings of the Fourteenth International Conference on Machine Learning (ICML'97), 1997.
[76] Rocchio J. Relevance Feedback in Information Retrieval [C]. In The Smart Retrieval System: Experiments in Automatic Document Processing, G. Salton, Ed. Prentice-Hall, Englewood Cliffs, NJ, 1971, 313-323.
[77] Belur V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques [M]. McGraw Hill Computer Science Series, IEEE Computer Society Press, Las Alamitos, California, 1996.
[78] Y. Yang and X. Liu. A Re-Examination of Text Categorization Methods [C]. Proceedings of ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR'99), 1999.
[79] Joachims, Thorsten. A Probabilistic Analysis of the Rocchio Algorithm with TFIDF for Text Categorization [C]. Proceedings of International Conference on Machine Learning (ICML), 1997.
[80] A. McCallum and K. Nigam. A Comparison of Event Models for Naive Bayes Text Classification [C]. In AAAI-98 Workshop on Learning for Text Categorization, 1998.
[81] Vapnik V. Nature of Statistical Learning Theory (2nd edition) [M]. New York, Springer Press, 2000.
[82] Joachims T. Text Categorization with Support Vector Machines: Learning with ManyRelevant Features [C]. Proceedings of 10th European Conference on Machine Learning(ECML-98), Chemnitz, DE, 1998, 137-142.
[83]史忠植.知识发现[M].北京:清华大学出版社, 2002年1月.
[84] Fran?ois Denis, Rémi Gilleronb, and Fabien Letouzeyb. Learning from Positive and Unlabeled Examples [J]. Theoretical Computer Science, Volume 348, Issue 1, 2 December 2005, Pages 70-83.
[85] Larry M. Manevitz, Malik Yousef. One-Class SVMs for Document Classification [J]. Journal of Machine Learning Research, 2001(12), 139-154.
[86] Hwanjo Yu, Jiawei Han, Kevin Chen-Chuan Chang. PEBL: Web Page Classification without Negative Examples [J]. IEEE Transactions on Knowledge and Data Engineering, 16(1): 70-81, January 2004.
[87] Hwanjo Yu. Single-Class Classification with Mapping Convergence[J]. Machine Learning, 61(1-3): 49-69, November 2005.
[88] Rakesh Agrawal, Roberto Bayardo, Ramakrishnan Srikant, Athena. Mining-based Interactive Management of Text Databases [C]. International Conference on Extending Database Technology, 2000, 365-379.
[89] Buckley, C., Salton G., and Allan J. The effect of adding relevance information in a relevance feedback environment [C]. SIGIR94, 1994.
[90] A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum Likelihood from Incomplete Data via the EM Algorithm [J]. J. Royal Stat. Soc., 39: 1-38, 1977.
[91] Hailong Yu, Wanli Zuo, Tao Peng. A New PU Learning Algorithm for Text Classification [C]. MICAI 2005: 824-832.
[92] Tao Peng, Wanli Zuo, Fengling He. SVM based Adaptive Learning Method for Text Classification from Positive and Unlabeled Documents [J]. Knowl. Inf. Syst. 16(3): 281-301, 2008.
[93] Hui Wang, and Wanli Zuo. Extracting Initial and Reliable Negative Documents to Enhance Classification Performance [C]. KDLL 2006, 104-111.
[94] Gabriel Pui Cheong Fung, Jeffrey Xu Yu, Hongjun Lu, Philip S. Yu. Text Classification without Labeled Negative Documents [C]. ICDE 2005, 594-605.
[95] Shuang Yu, Chunping Li. PE-PUC: A Graph based PU-Learning Approach for Text Classification [C]. 5th International Conference on Machine Learning and Data Mining in Pattern Recognition, Leipzig, Germany, Lecture Notes in Computer Science 4571, 574 -584, 2007.
[96] Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal. Semi-Supervised Learning on Directed Graphs [C]. Advances in Neural Information Processing Systems, 2005, 1633--1640.
[97] T. Joachims. A Support Vector Method for Multivariate Performance Measures [C]. Proceedings of the International Conference on Machine Learning (ICML), 2005.
[98] Xiaoli Li, and Bing Liu. Dealing with Different Distributions in Learning from Positive and Unlabeled Web Data [C]. WWW-2004.
[99] Xiaoli Li, and Bing Liu. Learning from Positive and Unlabeled Examples with Different DataDistributions [C]. European Conference on Machine Learning (ECML-05), Porto, Portugal, LNCS 3720, Springer, October 3-7, 2005, 218-229.
[100] Bing Liu, XiaoLi Li, W.S. Lee and P.S. Yu. Text Classification by Labeling Words [C]. Proceedings of Nineteenth National Conference on Artificial Intelligence (AAAI-2004), San Jose, California, USA. AAAI Press/MIT Press, July 25-29, 2004, 425-430.
[101] Gao Cong, Wee Sun Lee, Haoran Wu, Bing Liu. Semi-Supervised Text Classification Using Partitioned EM [C]. DASFAA 2004: 482-493.
[102] Borja Calvo, Paseo Manuel de Lardizabal, Pedro Larra?aga, Paseo Manuel de Lardizabal, JoséA. Lozano, Paseo Manuel de Lardizabal. Learning Bayesian Classifiers from Positive and Unlabeled Examples [J]. Pattern Recognition Letters archive, Volume 28, Issue 16, 2375-2384, 2007.
[103] Tao Peng, Fengling He, Wanli Zuo, Changli Zhang. Adaptive Topical Web Crawling for Domain-Specific Resource Discovery Guided by Link-Context [C]. MICAI 2006: 963-973.
[104] Xiaoli Li, Bing Liu, and S.-K. Ng. Learning to Identify Unexpected Instances in the Test Set [C]. Proceedings of Twenty International Joint Conference on Artificial Intelligence (IJCAI-07), Hyderabad, India, January 6-12, 2007, 2802-2807.
[105] Zhi-Hua Zhou. Learning with Unlabeled Data and Its Application to Image Retrieval [J]. Lecture Notes in Computer Science, Volume 4099/2006, Springer Berlin / Heidelberg, 2006, 5-10.
[106] Karl-Michael Schneider. Learning to Filter Junk E-Mail from Positive and Unlabeled Examples [J]. Lecture Notes in Computer Science, Springer Berlin / Heidelberg, Volume 3248/2005, 426-435.
[107] Wang C, Ding C, Meraz RF, Holbrook SR. PSoL: A Positive Sample Only Learning Algorithm for Finding Non-Coding RNA Genes [J]. Bioinformatics, 22(21): 2590-6, 2006.
[108] Charles Elkan, Keith Noto. Learning Classifiers from Only Positive and Unlabeled Data [C]. Proceeding of the 14th ACM SIGKDD, Las Vegas, Nevada, USA, 213-220, 2008.
[109]苏金树,张博锋,徐昕.基于机器学习的文本分类技术研究进展[J].软件学报, September 2006, 17(9): 1848?1859.
[110] Reuters-21578语料[EB/OL]. http://www.research.att.com/~lewis.
[111] Bow: A Toolkit for Statistical Language Modeling, Text Retrieval, Classification and Clustering [EB/OL]. http://www.cs.cmu.edu/~mccallum/bow/.
[112]周志华.半监督学习专刊前言[J].软件学报, 2008, 19(11): 2789?2790.
[113] D. J. Miller, H. S. Uyar. A Mixture of Experts Classifier with Learning based on both Labeled and Unlabelled Data [C]. In: M. Mozer, M. I. Jordan, T. Petsche, eds. Advances in Neural Information Processing Systems 9, Cambridge, MA: MIT Press, 1997, 571-577.
[114] T. Zhang, F. J. Oles. A Probability Analysis on the Value of Unlabeled Data for Classification Problems [C]. In: Proceedings of the 17th International Conference on Machine Learning (ICML’00), San Francisco, CA, 2000, 1191-1198.
[115] X. Zhu. Semi-Supervised Learning Literature Survey [R]. Technical Report 1530, Department of Computer Sciences, University of Wisconsin at Madison, Madison, WI, July19, 2008.
[116] Thorsten Joachims. Transductive Inference for Text Classification using Support Vector Machines[J]. Morgan Kaufmann, 1999, 200-209.
[117] Brendan J. Frey, Delbert Dueck. Clustering by Passing Messages Between Data Points, [J]. Science, 16 February 2007, Vol. 315, no. 5814, 2007, 972– 976.
[118] Kyriakopoulou, A. Text Classification Aided by Clustering: a Literature Review [M]. I-Tech Education and Publishing KG, Vienna, Austria (2008)
[119] Kyriakopoulou, A., Kalamboukis, T. Using Clustering to Enhance Text Classification [C]. In SIGIR 2007, 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 805–806, 2007.
[120] Kyriakopoulou, A., Kalamboukis, T. Text Classification using Clustering [C]. In: ECML/PKDD Discovery Challenge Workshop, 2006.
[121] Kyriakopoulou, A. Using Clustering and Co-Training to Boost Classification Performance [C]. In: ICTAI (2). IEEE Computer Society, 2007, 325–330.
[122] Tishby, N. Z., Pereira, F., Bialek, W. The Information Bottleneck Method [C]. In Proceedings of the 37th Allerton Conference on Communication, Control and Computing, 1999.
[123] Pereira F., Tishby N., Lee L. Distributional Clustering of English Words [C]. Proceedings of the 31st Annual Meeting of the Association for Computational Linguistics, 1993, 183-190.
[124] Baker L. D., McCallum A. K., (1998) Distributional Clustering of Words for Text Classification [C]. Proceedings of SIGIR’98, 21st ACM International Conference on Research and Development in Information Retrieval, pages 96–103, Melbourne, AU. ACM Press, New York, US.
[125] Dhillon I., Mallela S., Kumar R. A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification [J]. Journal of Machine Learning Research 3, 2003, 1265-1287.
[126] Slonim, N., Tishby, N. Document Clustering using Word Clusters via the Information Bottleneck Method[C]. In Proceedings of the ACM SIGIR, 2000.
[127] Yaniv R. E., Souroujon O. Iterative Double Clustering for Unsupervised and Semi-Supervised Learning [C]. In proceedings of the 12th European Conference on Machine Learning, ECML, 2001.
[128] J. Shi, J. Malik. Normalized Cuts and Image Segmentation [C]. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR'97), 731-737, 1997.
[129] Dhillon I. Co-Clustering Document and Word using Bipartite Spectral Graph Partitioning [C]. In: Proceedings of the ACM SIGKDD’01, 2001, 269-274.
[130] Dhillon I., Mallela S., Modha, S. Information Theoretic Co-Clustering [C] In Proceedings of the ACM SIGKDD Conference, 2003.
[131] Fung, G. and Mangasarian, O.L. Semi-Supervised Support Vector Machines for Unlabeled Data Classification [J]. Optim. Methods Software, 2001, 15(1): 29-44.
[132] Li, M., Cheng, Y., Zhao, H. Unlabeled Data Classification via Support Vector Machine andk-means Clustering[C]. In Proceedings of the International Conference on Computer Graphics, Imaging and Visualization, CGIV'04, 2004, 183-186.
[133] Hua-Jun Zeng, Xuan-Hui Wang, Zheng Chen, Hongjun Lu, Wei-Ying Ma. CBC: Clustering based Text Classification Requiring Minimal Labeled Data [C]. Third IEEE International Conference on Data Mining, 2003, Nov. 2003, 443– 450.
[134] Chapelle, O., Weston, J., Scholkopf, B. Cluster Kernels for Semi-Supervised Learning [C]. In NIPS, volume 15, 2002.
[135] Raskutti, B., Ferrá, H., Kowalczyk, A. Using Unlabelled Data for Text Classification through Addition of Cluster Parameters [C]. Proceedings of the Nineteenth International Conference on Machine Learning, 2000, 514–521.
[136] Raskutti, B., Ferrá, H., Kowalczyk, A. Combining Clustering and Co-Training to Enhance Text Classification using Unlabelled Data [C]. Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining, 2002.
[137] Yu, H., Yang, J., Han, J. Classifying Large Data Sets using SVMs with Hierarchical Clusters [C]. in Proceedings of the 9th ACM SIGKDD 2003, Washington, DC, USA, 2003.
[138] Boley, D., Cao, D. Training Support Vector Machine using Adaptive Clustering [C]. Proceeding of 2004 SIAM International Conference on Data Mining, 2004.
[139] Asharaf, S., Murty, M. N., Shevade, S. K. Cluster based Training for Scaling Nonlinear Support Vector Machines [C]. Proceedings of the International Conference on Computing: Theory and Applications (ICCTA'07), 2007.
[140] Li, B., Chi, M., Fan, J., Xue, X. Support Cluster Machine [C]. Proceedings of the 24th International Conference on Machine Learning, Corvallis, OR, 2007.
[141] Cervantes, J., Li, X., Yu, W., Li, K. Support Vector Machine Classification for Large Data Sets via Minimum Enclosing Ball Clustering [J]. Neurocomputing, 71(4-6): 611-619, 2008.
[142] Sun, S., Tseng, C. L., Chen, Y. H., Chuang, S. C., Fu, H. C. Cluster-based Support Vector Machines in Text-Independent Speaker Identification [C]. In Proceedings of the International Joint Conference on Neural Network, 2004.
[143] Wang, J., Wu, X., Zhang, C. Support Vector Machines based on k-Means Clustering for Real-Time Business Intelligence Systems [C]. Int. J. Business Intelligence and Data, Mining, 1(1): 54–64, 2005.
[144] He, J., Zhong, W., Harrison, R., Tai, P. C., Pan, Y. Clustering Support Vector Machines and Its Application to Local Protein Tertiary Structure Prediction [C]. ICCS 2006, part II, LNCS 3992, 2006, 710-717.
[145] CLUTO - Software for Clustering High-Dimensional Datasets [EB/OL]. http://glaros.dtc.umn.edu/ gkhome/ cluto/cluto/download.
[146] Thorsten Joachims. SVMlight Support Vector Machine [EB/OL]. http://svmlight.joachims.org/.
[147] W. Tang, H. Xiong, S. Zhong, and J. Wu. Enhancing Semi-Supervised Clustering: A Feature Projection Perspective [C]. Proceedings of the 13th ACM SIGKDD, 2007, 707-716.
[148] Demiriz A, Bennett KP, Embrechts MJ. Semi-Supervised Clustering using GeneticAlgorithms [C]. In: Dagli CH, ed. Proc. of the Intelligent Engineering Systems Through Artificial Neural Networks 9 (ANNIE’99). New York: ASME Press, 1999, 809?814.
[149] Wagstaff K, Cardie C, Rogers S, Schroedl S. Constrained K-means Clustering with Background Knowledge [C]. In: Carla EB, Andrea PD, eds. Proc. of the 18th Int’l Conf. on Machine Learning (ICML 2001). San Fransisco: Morgan Kaufmann Publishers, 2001, 577?584.
[150] Basu S, Bilenko M, Mooney RJ. A Probabilistic Framework for Semi-Supervised Clustering [C]. In: Won K, Ron K, Johannes G William D, eds. Proc. of the 10th ACM SIGKDD Int’l Conf. on Knowledge Discovery and Data Mining (KDD 2004). Seattle: ACM Press, 2004, 59?68.
[151] Basu S, Banerjee A, Mooney RJ. Semi-Supervised Clustering by Seeding [C]. In: Claude S, Achim GH, eds. Proc. of the 19th Int’l Conf. on Machine Learning (ICML 2002), San Fransisco: Morgan Kaufmann Publishers, 2002, 19?26.
[152] Basu S. Semi-Supervised clustering: Probabilistic Models Algorithms and Experiments [D]. Austin: University of Texas at Austin, 2005.
[153] Xing EP, Ng AY, Jordan MI, Russell S. Distance Metric Learning with Application to Clustering with Side-information [C]. In: Becher S, Thrun S, Obermayer K, eds. Proc. of the 16th Annual Conf. on Neural Information Processing System, Cambridge: MIT Press, 2003, 505?512.
[154] Bar-Hillel A, Hertz T, Shental N, Weinshall D. Learning Distance Functions using Equivalence Relations [C]. In: Fawcett T, Mishra N, eds. Proc. of the 20th Int’l Conf. on Machine Learning, Washington: Morgan Kaufmann Publishers, 2003, 11?18.
[155] Yeung DY, Chang H. Extending the Relevant Component Analysis Algorithm for Metric Learning using both Positive and Negative Equivalence Constraints [J]. Pattern Recognition, 2006, 39(5):1007?1010.
[156] Schultz M, Joachims T. Learning a Distance Metric from Relative Comparisons [C]. In: Thrun S, Saul LK, Sch?lkopf B, eds. Proc. of the 17th Annual Conf. on Neural Information Processing System, Cambridge: MIT Press, 2004, 41?48.
[157] De la Torre F, Kanade T. Discriminative Cluster Analysis [C]. In: William WC, Andrew M, eds. Proc. of the 19th Int’l Conf. on Machine Learning, New York: ACM Press, 2006, 241?248.
[158] Ding CH, Li T. Adaptive Dimension Reduction using Discriminant Analysis and K-means Clustering [C]. In: Ghahramani Z, ed. Proc. of the 19th Int’l Conf. on Machine Learning, New York: ACM Press, 2007, 521?528.
[159] Ye JP, Zhao Z, Liu H. Adaptive Distance Metric Learning for Clustering [C]. In: Bishop CM, Frey B, eds. Proc. of IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, Madison: IEEE Computer Society Press, 2007, 1?7.
[160]尹学松,胡恩良,陈松灿.基于成对约束的判别型半监督聚类分析[C].软件学报, 2008, 19(11): 2791?2802.
[161] Bilenko M, Basu S, Mooney RJ. Integrating Constraints and Metric Learning inSemi-Supervised Clustering [C]. In: Brodley CE, ed. Proc. of the 21st Int’l Conf. on Machine Learning, New York: ACM Press, 2004, 81?88.
[162] Basu S, Banerjee A, Mooney RJ. A Probabilistic Framework for Semi-Supervised Clustering [C]. In: Boulicaut JF, Esposito F, Giannotti F, Pedreschi D, eds. Proc. of the 10th ACM SIGKDD Int’l Conf. on Knowledge Discovery and Data Mining, New York: ACM Press, 2004, 59?68.
[163] Brian Kulis, Sugato Basu, Inderjit Dhillon, and Raymond Mooney. Semi-Supervised Graph Clustering: A Kernel Approach [J]. Machine Learning Journal, 74(1): 1-22, 2009.
[164] Foster Provost, Pedro Domingos. Tree Induction for Probability-based Ranking, [J]. Machine Learning, 2003, 199-215.
[165] Maytal Saar-tsechansky, Foster Provost. Active Sampling for Class Probability Estimation and Ranking [J]. Machine Learning, 2004, 153-178.
[166] Chris Burges, Tal Shaked, Erin Renshaw, Ari Lazier, Matt Deeds, Nicole Hamilton, Greg Hullender. Learning to Rank using Gradient Descent [C]. in Proceedings of the 22nd International Conference on Machine Learning, Bonn, Germany, 2005.
[167] Pinar Donmez, Jaime G. Carbonell. Optimizing Estimated Loss Reduction for Active Sampling in Rank Learning: DiffLoss [C]. ICML 2008, Helsinki, Finland, June 7, 2008.
[168] J. Xu and H. Li. Adarank: A Boosting Algorithm for Information Retrieval [C]. In SIGIR 2007, 391-398, 2007.
[169] Hroza, J., ?i?ka, J. Selecting Interesting Articles Using Their Similarity Based Only on Positive Examples [C]. Proceedings of the Sixth International Conference on Intelligent Text Processing and Computational Linguistics CICLing-2005, February 13-19, 2005, Mexico City, Mexico, Springer-Verlag, 2005, LNCS/LNAI 3406, 608- 611.
[170] Kevin Duh, Katrin Kirchhoff. Learning to Rank with Partially-Labeled Data [C]. Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval, Singapore, 2008, 251-258.
[171] Y. Yang. An Evaluation of Statistical Approaches to Text Categorization, Information Retrieval archive [M]. Kluwer Academic Publishers, Hingham, MA, USA, 1(1-2): 69-90, 1999.
[172] Kamal Nigam, Rayid Ghani. Analyzing the Effectiveness and Applicability of Co-Training [C]. In Proc. of Ninth International Conference on Information and Knowledge (CIKM-2000).
[173] Pierce, David and Cardie, Claire. Limitations of Co-Training for Natural Language Learning from Large Datasets [C]. Proceedings of the 2001 Conference on Empirical Methods in Natural Language Processing, 2001, 1-9.
[174] S. Dasgupta, M. Littman, D. McAllester. PAC Generalization Bounds for Co-Training [C]. In: T. G. Dietterich, S. Becker, Z. Ghahramani, eds. Advances in Neural Information Processing Systems 14, Cambridge, MA: MIT Press, 2002, 375-382.
[175] M.-F. Balcan, A. Blum, K. Yang. Co-Training and Expansion: Towards Bridging Theory and Practice [C]. In: L. K. Saul, Y. Weiss, L. Bottou, eds. Advances in Neural Information Processing Systems 17, Cambridge, MA: MIT Press, 2005, 89-96.
[176] W. Wang, Z.-H. Zhou. Analyzing Co-Training Style Algorithms [C]. In: Proceedings of the 18th European Conference on Machine Learning (ECML’07), Warsaw, Poland, 2007.
[177] Z.-H. Zhou, M. Li. Semi-Supervised Learning with Co-Training [C]. In: Proceedings of the 19th International Joint Conference on Artificial Intelligence (IJCAI’05), Edinburgh, Scotland, 2005, 908-913.
[178] Z.-H. Zhou, M. Li. Semi-Supervised Learning with Co-Training Style Algorithm [J]. IEEE Transactions on Knowledge and Data Engineering, 19(11), 2007.
[179] U. Brefeld, T. G?rtner, T. Scheffer, S. Wrobel. Efficient Co-Regularised Least Squares Regression [C]. In: Proceedings of the 23rd International Conference on Machine Learning (ICML’06), Pittsburgh, PA, 2006, 137-144.
[180] D. Yarowsky. Unsupervised Word Sense Disambiguation Rivaling Supervised Methods [C]. In: Proceedings of the 33rd Annual Meeting of the Association for Computational Linguistics (ACL’95), Cambridge, MA, 1995, 189-196.
[181] S., Kiritchenko, S. Matwin. Email Classification with Co-training [R]. (Technical Report), University of Ottawa, 2002.
[182] Jason Chan, Irena Koprinska, Josiah Poon. Co-training with a Single Natural Feature Set Applied to Email Classification [C]. Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence, 2004, 586-589.
[183]张博锋,苏金树,徐昕.文本分类中用于协同训练的特征集分割[J/OL].中国科技论文在线, http://www.edu.cn, 2008-04-01.
[184] Ion Muslea, Steven Minton, Craig A. Knoblock. Adaptive View Validation: A First Step Towards Automatic View Detection [C]. Proceedings of ICML 2002, 443-450.
[185] M. Li, Z.-H. Zhou. SETRED: Self-Training with Editing [C]. Proceedings of 9th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD’05), Hanoi, Vietnam, LNAI 3518, 2005, 611-621.
[186] Ion Muslea, Steve Minton, Craig Knoblock. Selective Sampling with Redundant Views, [C]. Proceedings of the Fifteenth National Conference on Artificial Intelligence, AAAI-2000, 621-626.
[187] Ulf Brefeld, Tobias Scheffer. Co-EM Support Vector Learning [C]. Proceedings of the twenty-first international conference on Machine learning, Banff, Alberta, Canada, 2004.
[188] S. Goldman, Y. Zhou. Enhancing Supervised Learning with Unlabeled Data [C]. In: Proceedings of the 17th International Conference on Machine Learning (ICML’00), San Francisco, CA, 2000, 327-334.
[189] D. Angluin, P. Laird. Learning from Noisy Examples [J]. Machine Learning, 1988, 2(4): 343-370.
[190] Y. Zhou, S. Goldman. Democratic Co-learning [C]. In: Proceedings of the 16th IEEE International Conference on Tools with Artificial Intelligence (ICTAI’04), Boca Raton, FL, 2004, 594–602.
[191] D. Mavroeidis, K. Chaidos, S. Pirillos, D. Christopoulos, M. Vazirgiannis. Using Tri-Training and Support Vector Machines for Addressing the ECML/PKDD 2006 DiscoveryChallenge [C]. In: Proceedings of the ECML-PKDD Discovery Challenge Workshop, Berlin, Germany, 2006, 39-47.
[192] N. T. Tri, N. M. Le, and Akira S. Using Semi-Supervised Learning for Question Classification [C]. Lecture Notes in Computer Science, Volume 4285, 2006, 31-41.
[193] W. L. Chen, Y. J. Zhang, and Isahara H. Chinese Chunking with Tri-Training Learning [C]. 21st International Conference on the Computer Processing of Oriental Languages (ICCPOL2006), 2006, 466-473.
[194]邓超,郭茂祖.基于自适应数据剪辑策略的Tri-training算法[J].计算机学报, 2007年8月, 30(8): 1213-1226.
[195]邓超,郭茂祖.基于Tri-training和数据剪辑的半监督聚类算法[J].软件学报, 2008, 19(3): 663-673.
[196] M. Li, Z.-H. Zhou. Improve Computer-aided Diagnosis with Machine Learning Techniques using Undiagnosed Samples [J]. IEEE Transactions on Systems, Man and Cybernetics - Part A: Systems and Humans, 2007, 37(6): 1088-1098.