基于Tucker分解的半监督支持张量机

doi:10.11896/j.issn.1002-137X.2019.09.028

摘要/Abstract

摘要： 传统的机器学习方法所使用的数据大多是基于向量空间的。支持向量机(Support Vector Machine,SVM)作为一种重要的机器学习方法,在解决小样本、非线性、高维数据等问题时具有较好的性能。但在实际应用中,图像和视频等数据都是用张量形式表示的,如果将这些张量数据直接转换成向量数据,往往会丢失一些原有的结构和相关性信息,有可能造成维度灾难和小样本问题。为了设法保持尽量多的张量结构信息,提出了一种采用Tucker分解的支持张量机(Support Tensor Machine,STM)算法。实验表明,该方法可以明显提高分类器性能;同时,支持张量机作为监督学习方式,存在无法利用未标记数据的缺点,往往受限于训练数据不足的情况。因此,将半监督学习方法与支持张量机相结合,进而提出了基于Tucker分解的半监督支持张量机算法(Semi-Supervised STM,S3TM)。该算法既可以保持较多的张量结构信息,又能充分利用未标记数据。实验表明,采用该算法的预测准确率达到90.26%,从而验证了所提算法的有效性。

关键词: Tucker分解, 半监督学习, 支持张量机

Abstract: Most data used by traditional machine learning methods belong to vector space.As an important machine learning method,support vector machine (SVM) has better performance in solving small samples,nonlinearity and high-dimensionality problems.However,in practical applications,some data like images and videos are both stored in tensor form.If tensor data are convert into vector data,some original structure and related information may be lost,which will cause dimensional disasters and small samples problems.Therefore,to maintain as much tensor structure information as possible,support tensor machine (STM) based on Tucker decomposition was proposed.Experiments show that this method can significantly improve the classification performance.Meanwhile,as a supervised learning method,support vector machine cannot use unlabeled data,which often encounter problems with insufficient training data.Therefore,semi-supervised support tensor machine based on Tucker decomposition was proposed.This algorithm can not only maintain more tensor structure information,but also make full use of the unlabeled data.Experiments show that the prediction accuracy rate is 90.26%,which validates the effectiveness of the proposed method.

Key words: Semi-supervised learning, Support tensor machine, Tucker decomposition

中图分类号:

TP181

吴振宇, 李云雷, 吴凡. 基于Tucker分解的半监督支持张量机[J]. 计算机科学, 2019, 46(9): 195-200. https://doi.org/10.11896/j.issn.1002-137X.2019.09.028

WU Zhen-yu, LI Yun-lei, WU Fan. Semi-supervised Support Tensor Based on Tucker Decomposition[J]. Computer Science, 2019, 46(9): 195-200. https://doi.org/10.11896/j.issn.1002-137X.2019.09.028

参考文献

[1]GUO X,HUANG X,ZHANG L,et al.Support Tensor Machines for Classification of Hyperspectral Remote Sensing Imagery[J].IEEE Transactions on Geoscience & Remote Sen-sing,2016,54(6):3248-3264.
[2]ZHANG J,HAN Y,JIANG J.Tucker decomposition-based ten-sor learning for human action recognition[J].Multimedia Systems,2016,22(3):343-353.
[3]RAJWADE A,RANGARAJAN A,BANERJEE A.Image De-noising Using the Higher Order Singular Value Decomposition[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2013,35(4):849-862.
[4]BISWAS S K,MILANFAR P.Linear Support Tensor Machine with LSK Channels: Pedestrian Detection in Thermal Infrared Images[J].IEEE Transactions on Image Processing,2017,26(9):4229-4242.
[5]CAI D,HE X F,WEN J R,et al.Support Tensor Machines for Text Categorization[J].International Journal of Academic Research in Business & Social Sciences,2006,2(12):2222-6990.
[6]LUO L,XIE Y,ZHANG Z,et al.Support matrix machines[C]//International Conference on International Conference on Machine Learning.JMLR.org,2015:938-947.
[7]TAO D,LI X,HU W,et al.Supervised Tensor Learning[C]//IEEE International Conference on Data Mining.IEEE,2005.
[8]KOTSIA I,PATRAS I.Support tucker machines[C]//The24th IEEE Conference on Computer Vision and Pattern Recognition(CVPR 2011).Colorado Springs,CO,USA,IEEE,2011.
[9]HAO Z,HE L,CHEN B,et al.A Linear Support Higher-Order Tensor Machine for Classification[J].IEEE Transactions on Image Processing,2013,22(7):2911-2920.
[10]CHEN Y,LU L,ZHONG P.One-class support higher ordertensor machine classifier[J].Applied Intelligence,2017,47(7):1-9.
[11]CHEN Z,BATSELIER K,SUYKENS J A K,et al.Parallelized Tensor Train Learning of Polynomial Classifiers[J].IEEE Transactions on Neural Networks & Learning Systems,2016,PP(99):1-12.
[12]SHU T.Research on Regression Problem and ClassificationProblem Algorithm with Input Tensor Data[D].Urumqi :Xinjiang University,2017.(in Chinese)舒托.输入为张量的回归问题和分类问题算法研究[D].乌鲁木齐:新疆大学,2017.
[13]YU J,LIU G S.Text Classification Research Based on Support Tensor[J].Information Technology,2016(9):10-13.(in Chinese)俞炯,刘功申.基于支持张量机的文本分类研究[J].信息技术,2016(9):10-13.
[14]LIU J W,LIU Y,LUO X L.Semi-supervised learning algorithm[J].Chinese Journal of Computers,2015,38(8):1592-1617.(in Chinese)刘建伟,刘媛,罗雄麟.半监督学习方法[J].计算机学报,2015,38(8):1592-1617.
[15]CHEN Y H.Research on Semi-supervised learning classification algorithm[D].Zhenjiang:Jiangsu University,2017.(in Chinese)陈英豪.半监督学习分类算法的研究[D].镇江:江苏大学,2017.
[16]WIDMANN N,VERBERNE S.Graph-based Semi-supervisedLearning for Text Classification[C]//Acm Sigir International Conference on Theory of Information Retrieval.ACM,2017.
[17]HADY M F A,SCHWENKER F.Semi-supervised Learning[J].Intelligent Systems Reference Library,2013,49(2):215-239.
[18]XU C,TAO W.Combining Active Learning and Semi-Super-vised Learning by Using Selective Label Spreading [C]//IEEE International Conference on Data Mining Workshops.IEEE,2017.
[19]TUCKER L R.Extension of factor analysis to three-dimensional matrices[M]// Contributions to Mathematical Psychology.New York:Helt Rinehart & Winston,1964.
[20]TUCKER L R.Some mathematical notes on three-mode factor analysis[J].Psychometrika,1966,31(3):279-311.
[21]YANG B.Research and Application of Machine Learning Method Based on Tensor Data[D].Beijing:China Agricultural University,2014.(in Chinese)杨兵.基于张量数据的机器学习方法研究与应用[D].北京: 中国农业大学,2014.
[22]ERFANI S M,BAKTASHMOTLAGH M,RAJASEGARAR S,et al.R1STM:One-class Support Tensor Machine with Randomised Kernel[C]//SIAM International Conference on Data Mining.Philadelphia:Society for Industrial and Applied Mathematics Conference,2016.

相关文章 15

[1]	武红鑫, 韩萌, 陈志强, 张喜龙, 李慕航. 监督和半监督学习下的多标签分类综述 Survey of Multi-label Classification Based on Supervised and Semi-supervised Learning 计算机科学, 2022, 49(8): 12-25. https://doi.org/10.11896/jsjkx.210700111
[2]	侯夏晔, 陈海燕, 张兵, 袁立罡, 贾亦真. 一种基于支持向量机的主动度量学习算法 Active Metric Learning Based on Support Vector Machines 计算机科学, 2022, 49(6A): 113-118. https://doi.org/10.11896/jsjkx.210500034
[3]	庞兴龙, 朱国胜. 基于半监督学习的网络流量分析研究 Survey of Network Traffic Analysis Based on Semi Supervised Learning 计算机科学, 2022, 49(6A): 544-554. https://doi.org/10.11896/jsjkx.210600131
[4]	王宇飞, 陈文. 基于DECORATE集成学习与置信度评估的Tri-training算法 Tri-training Algorithm Based on DECORATE Ensemble Learning and Credibility Assessment 计算机科学, 2022, 49(6): 127-133. https://doi.org/10.11896/jsjkx.211100043
[5]	许华杰, 陈育, 杨洋, 秦远卓. 基于混合样本自动数据增强技术的半监督学习方法 Semi-supervised Learning Method Based on Automated Mixed Sample Data Augmentation Techniques 计算机科学, 2022, 49(3): 288-293. https://doi.org/10.11896/jsjkx.210100156
[6]	杨洋, 邸一得, 刘俊晖, 易超, 周维. 基于张量分解的排序学习在个性化标签推荐中的研究 Study on Learning to Rank Based on Tensor Decomposition in Personalized Tag Recommendation 计算机科学, 2020, 47(11A): 515-519. https://doi.org/10.11896/jsjkx.191100181
[7]	秦悦, 丁世飞. 半监督聚类综述 Survey of Semi-supervised Clustering 计算机科学, 2019, 46(9): 15-21. https://doi.org/10.11896/j.issn.1002-137X.2019.09.002
[8]	沈鸿, 刘军发, 陈益强, 蒋鑫龙, 黄正宇. 基于多模融合的半监督场景识别方法 Semi-supervised Scene Recognition Method Based on Multi-mode Fusion 计算机科学, 2019, 46(12): 306-312. https://doi.org/10.11896/jsjkx.191200500C
[9]	喻影, 陈珂, 寿黎但, 陈刚, 吴晓凡. 基于关键词和关键句抽取的用户评论情感分析 Sentiment Analysis of User Comments Based on Extraction of Key Words and Key Sentences 计算机科学, 2019, 46(10): 19-26. https://doi.org/10.11896/jsjkx.191000531C
[10]	刘枭, 王晓国. 基于概率图的银行电信诈骗检测方法 Probabilistic Graphical Model Based Approach for Bank Telecommunication Fraud Detection 计算机科学, 2018, 45(7): 122-128. https://doi.org/10.11896／j.issn.1002-137X.2018.07.020
[11]	蔡毅,朱秀芳,孙章丽,陈阿娇. 半监督集成学习综述 Semi-supervised and Ensemble Learning:A Review 计算机科学, 2017, 44(Z6): 7-13. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.002
[12]	成英超,王瑞胡,胡章平. 一种基于高斯混合模型的协同过滤算法 Novel Approach on Collaborative Filtering Based on Gaussian Mixture Model 计算机科学, 2017, 44(Z6): 451-454. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.101
[13]	李锋,万小强. 基于关联矩阵的短信自动分类 SMS Automatic Classification Based on Relational Matrix 计算机科学, 2017, 44(Z6): 428-432. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.096
[14]	张维,苗夺谦,高灿,李峰. 一种处理部分标记数据的粗糙集属性约简算法 Rough Set Attribute Reduction Algorithm for Partially Labeled Data 计算机科学, 2017, 44(1): 25-31. https://doi.org/10.11896/j.issn.1002-137X.2017.01.005
[15]	朱常宝,程勇,高强. 基于半监督深度信念网络的图像分类算法研究 Research on Image Classification Algorithm Based on Semi-supervised Deep Belief Network 计算机科学, 2016, 43(Z6): 46-50. https://doi.org/10.11896/j.issn.1002-137X.2016.6A.010

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed