基于半监督和迁移学习的近红外光谱建模方法研究
|
摘要
在科学与经济飞速发展的当今时代，企业生产过程的自动化和智能化水平日益提高，传统的产品质量监控手段难以满足产品研发和生产控制的需求。近红外光谱分析技术(NIR)作为一种新型、快速高效的检测手段应运而生，大大提升了产品质量监督管理的工作效率，已经在石油、医药、烟草等行业中被广泛应用。
通过总结前期在“智能感官评估方法”课题的研究结论得知，当模型的输入是信息不够完备的常规化学成分指标时，难以建立分类性能良好的感官模拟评估模型。传统的实验室化学分析法往往检测的成分数量有限，而近红外光谱中包含了丰富的成分信息。近红外光谱分析通常应用于检测产品的化学成分含量方面，本文以近红外光谱作为研究对象，应用多种机器学习方法，深入分析近红外光谱中反映的卷烟产品质量以及卷烟配方中成分的关系。直接应用高维的光谱数据建立与产品质量之间的关系模型。
近年的实践应用表明，传统近红外光谱分析技术在面对成分复杂或背景噪声干扰较大的情况时，遇到了模型稳定性差、预测误差较大、建模样本量大、模型难以移植等许多实际问题，现有的近红外光谱建模技术亟待提升。本文从近红外光谱分析建模的基本原理、国内外研究现状分析入手，在转导推理思想的启发下，将半监督学习、迁移学习方法引入近红外光谱分析建模方法体系，主要围绕近红外高维光谱数据处理、光谱定性分析和定量分析建模、光谱分析模型传递四个关键技术内容展开深入研究。论文的主要研究内容为：
1)当近红外光谱与观测数据为非线性关系时，传统降维方法容易出现原始数据特征信息丢失、流形结构被破坏、数据分类性能下降等问题。本文提出一种半监督核邻域保护嵌入算法(SSKNPE)。该算法基于核变换距离将非线性问题转化为一个特征空间的线性问题，通过充分利用部分有标记样本的先验分类信息约束特征映射，使数据从高维映射到低维后仍能保持数据的全局结构和局部结构。实验验证，SSKNPE算法的降维质量优于LLE等传统流形学习算法，能更好地改善卷烟品牌识别近红外光谱分析模型的分类性能。
2)针对传统分类器的归纳推理机制存在的预测风险问题和大量有标记的训练样本的约束等问题，引入转导推理和半监督学习思想，提出了一种基于近邻传播聚类的半监督支持向量机算法(APS4VM)。算法将近邻传播聚类和混沌优化相结合，快速搜索多个间隔最大平面的低密度区域，确定安全分类的支持决策面。算法在少量标记样本的情况下，针对Iris数据集和卷烟口味评价数据进行实验验证。实验结果表明能够建立了性能良好、稳健的分类模型，半监督支持向量机具有实际工程应用价值，解决了标记样本不足时的卷烟近红外光谱定性分析建模困难的问题。
3)针对复杂非线性问题中传统近红外光谱定量建模方法预测性能较低，要求训练样本足够多等实际工程应用困难，提出一种基于量子粒子群优化的半监督支持向量回归算法(QPSO-LSS3VR)。该算法结合K近邻和置信度选样方法实现半监督学习中未标记样本估计，采用高效的量子粒子群优化算法搜索最佳的半监督支持向量机回归模型参数γ, λ,σ。卷烟总糖预测模型实验表明，该算法基于半监督学习思想，能在少量标记样本的情况下快速达到较低的预测标准误差，优化方法提高了建模的时间效率，同时降低了半监督建模成本，解决了标记样本不足时的卷烟近红外光谱定量分析建模困难的问题。
4)针对近红外光谱分析仪器之间模型通用性差的问题，分析现有模型传递方法的不适用性：建模所需的标准样本准备条件苛刻，实际操作复杂，传统统计方法传递后的模型预测性能偏低。本文创新性地应用迁移学习思想，提出一种新的近红外光谱模型传递算法，即基于相似匹配和迁移学习的模型传递算法(SM-TrBoostEns)。通过非线性降维方法将近红外光谱投影到低维空间，根据距离度量样品的相似性，筛选对目标仪器建模有益的样本进行知识迁移，并采用迁移式Boosting技术和集成学习相结合的方式传递模型。通过两台近红外设备之间的卷烟总糖预测模型传递实验表明，该算法在目标仪器采集较少标准样本光谱的情况下，仍能有效提升目标仪器回归预测精度，具有一定的实用性。实验同时也说明迁移学习方法可以在近红外光谱模型传递方面继续深入探索和改进。
5)总结本文的研究结论和创新工作，提出下一步研究工作重点将围绕半监督学习模型的预测输出置信度、异常光谱凸壳判别、特征波长筛选等方面开展研究，逐步建立起基于近红外光谱分析的产品质量评价等应用的技术框架。
In today's era, the science and economy have developed rapidly. The level of theproduction process is becoming more and more automatic and intelligent. Thetraditional means of product quality control can not meet the needs of productdevelopment and control. Near-infrared spectroscopy (NIR) has emerged as a newdetection technology with fast and efficient features. It can greatly improve theefficiency of product quality supervision and management, and now it is widely usedin the oil, medicine, tobacco industry, and etc.
By summarizing the conclusion of early subject named as “Intelligent Methodsof Sensory Evaluation”, it shows that building a sensory evaluation model with goodclassification performance is difficult when the input of the model is chemicalcomposition with incomplete information. The number of components is oftenlimited by detecting with traditional methods of chemical analysis, but near-infraredspectral contains a wealth of component information. The near-infrared spectroscopyalways applied to detect the chemical components of products. In this paper, kinds ofmachine learning methods are applied in the research field of near-infraredspectroscopy. The features of cigarette quality and the components of cigaretteblends is mined from the near-infrared spectroscopy. The high-dimensional spectraldata is directly applied to build the relationships model between product quality andspectra.
Many practices show when NIR is applied to resolve the prediction problem ofcomplex component under the environment with large background noise, thetraditional analysis technology of near-infrared spectroscopy will always need a largeamount of samples, and the model always has poor performance and stability, and it isdifficult to transfer. So the existing modeling technology of NIR should be improved.This paper begins with analyzing the basic principles of modeling of NIR indomesticand aboard research status, semi-supervised learning and transfer learning are introduced for modeling method system of NIR under the inspiration of transductivereference. The paper describes the four key technologies about the high-dimensionaldata processing of NIR, qualitative analysis of spectral, quantitative analysis ofspectral and model transfer. The main content of this paper is as:
1) When the relationship between the observation data and near infrared spectra(NIR) is nonlinear, traditional dimension reduction methods can easily cause to losecharacteristic information, destroy manifold structure and decline the performance ofclassifier and so on. In this paper, a novel algorithm based on semi-supervised kernelneighborhood protection embedding is provided, named as SSKNPE. The kenerldistance is used to transform nonlinear problem into the linear problem in newfeatures space. It can take advantage of the prior classified information from labeleddata, and constraint feature mapping, so the data is mapped from high-dimensionspace to low-dimension space with preserving global structure and local structure.The experimental results show that SSKNPE can effectively improve theclassification ability after dimension reduction. The SSKNPE algorithm is applied tosolve the the cigarette brand identification problem based on the near-infraredspectrum.
2) Because of the inductive reference mechanism, the traditional classifier has theproblem of large prediction risk and training sample number. In this paper, weintroduce transduction reference and semi-supervised learning, and provide a novelsemi-supervised and support vector machines based on affinity propagation clustering(APS4VM). The low density area is found from many large margins by combiningaffinity propagation clustering with chaos optimization. The method can find asupport vector decision surface which can classify samples safely. Though there arefew labeled data in the iris data set and the taste evaluation data set of cigarettes, thegood performance of classifier is obtained based APS4VM algorithm. Sosemi-supervised support vector machine has practical value in engineering application.It is suitable for building the qualitative model of cigarette taste evaluation.
3) The quantitative analysis model based on traditional regression method can notperform well when it meets complex nonlinear problems, particularly when the training samples are not enough. In this paper a novel semi-supervised support vectorregression algorithm (QPSO-LSS3VR) is provided which is based on quantumparticle swarm optimization. The unlabeled samples are estimated by combining theK-nearest neighbor and confidence selection method. The best parameters(γ, λ,σ) ofthe semi-supervised support vector regression model are found by QPSO algorithm.The experiment result of predicting total sugar of cigarette shows that this algorithmcan effectively reduce the standard error of prediction and the cost of modelling whenthere are few training samples, and the algorithm can be applied to build the model ofsugar content prediction of cigarette.
4) The problem is that model versatility between near-infraed spectrometers isbad. The existing model transfer methods are analyzed its non-applicability: thepreparation condition of standard samples for modeling are required harshly, and thepractice operation process is complex, and the model has low prediction performanceafter using traditional statistical model transfer methods.In this paper, a new modeltransfer algorithm of near infrared spectral is innovately provided. It includes the ideaof transfer learning and similarity sample distance metric, named as SM-TrBoostEns.NIR spectral is projected into a low-dimensional space by nonlinear dimensionreduction method. According to metic the similarity between samples, the knowledgetransfers by that useful samples are selected for modeling on the target instrument,and the model is transferred with the combination of transfer boosting technology andensemble learning. The experiments of predicting cigarette total sugar by transferringmodel between two NIR instruments shows that the algorithm can still effectivelyincrease the regression precision under the condition that less standard samples’spectrals are collected on target instrument, so this algorithm has a certain practical.The experiments also shows that transfer learning can be explored and improved indepth for applying to the model transfer of NIR.
5) At last, the research conclusion and innovation is summarized. In the future,more relevant content will be researched, such as the output confidence of semi-supervised learning model, abnormal spectral sample discrimination based onconvex hull, and features wavelength filtering and so on. All results will help us tobuild a new framework of quality evaluation of product based on the near infraredspectra.
通过总结前期在“智能感官评估方法”课题的研究结论得知，当模型的输入是信息不够完备的常规化学成分指标时，难以建立分类性能良好的感官模拟评估模型。传统的实验室化学分析法往往检测的成分数量有限，而近红外光谱中包含了丰富的成分信息。近红外光谱分析通常应用于检测产品的化学成分含量方面，本文以近红外光谱作为研究对象，应用多种机器学习方法，深入分析近红外光谱中反映的卷烟产品质量以及卷烟配方中成分的关系。直接应用高维的光谱数据建立与产品质量之间的关系模型。
近年的实践应用表明，传统近红外光谱分析技术在面对成分复杂或背景噪声干扰较大的情况时，遇到了模型稳定性差、预测误差较大、建模样本量大、模型难以移植等许多实际问题，现有的近红外光谱建模技术亟待提升。本文从近红外光谱分析建模的基本原理、国内外研究现状分析入手，在转导推理思想的启发下，将半监督学习、迁移学习方法引入近红外光谱分析建模方法体系，主要围绕近红外高维光谱数据处理、光谱定性分析和定量分析建模、光谱分析模型传递四个关键技术内容展开深入研究。论文的主要研究内容为：
1)当近红外光谱与观测数据为非线性关系时，传统降维方法容易出现原始数据特征信息丢失、流形结构被破坏、数据分类性能下降等问题。本文提出一种半监督核邻域保护嵌入算法(SSKNPE)。该算法基于核变换距离将非线性问题转化为一个特征空间的线性问题，通过充分利用部分有标记样本的先验分类信息约束特征映射，使数据从高维映射到低维后仍能保持数据的全局结构和局部结构。实验验证，SSKNPE算法的降维质量优于LLE等传统流形学习算法，能更好地改善卷烟品牌识别近红外光谱分析模型的分类性能。
2)针对传统分类器的归纳推理机制存在的预测风险问题和大量有标记的训练样本的约束等问题，引入转导推理和半监督学习思想，提出了一种基于近邻传播聚类的半监督支持向量机算法(APS4VM)。算法将近邻传播聚类和混沌优化相结合，快速搜索多个间隔最大平面的低密度区域，确定安全分类的支持决策面。算法在少量标记样本的情况下，针对Iris数据集和卷烟口味评价数据进行实验验证。实验结果表明能够建立了性能良好、稳健的分类模型，半监督支持向量机具有实际工程应用价值，解决了标记样本不足时的卷烟近红外光谱定性分析建模困难的问题。
3)针对复杂非线性问题中传统近红外光谱定量建模方法预测性能较低，要求训练样本足够多等实际工程应用困难，提出一种基于量子粒子群优化的半监督支持向量回归算法(QPSO-LSS3VR)。该算法结合K近邻和置信度选样方法实现半监督学习中未标记样本估计，采用高效的量子粒子群优化算法搜索最佳的半监督支持向量机回归模型参数γ, λ,σ。卷烟总糖预测模型实验表明，该算法基于半监督学习思想，能在少量标记样本的情况下快速达到较低的预测标准误差，优化方法提高了建模的时间效率，同时降低了半监督建模成本，解决了标记样本不足时的卷烟近红外光谱定量分析建模困难的问题。
4)针对近红外光谱分析仪器之间模型通用性差的问题，分析现有模型传递方法的不适用性：建模所需的标准样本准备条件苛刻，实际操作复杂，传统统计方法传递后的模型预测性能偏低。本文创新性地应用迁移学习思想，提出一种新的近红外光谱模型传递算法，即基于相似匹配和迁移学习的模型传递算法(SM-TrBoostEns)。通过非线性降维方法将近红外光谱投影到低维空间，根据距离度量样品的相似性，筛选对目标仪器建模有益的样本进行知识迁移，并采用迁移式Boosting技术和集成学习相结合的方式传递模型。通过两台近红外设备之间的卷烟总糖预测模型传递实验表明，该算法在目标仪器采集较少标准样本光谱的情况下，仍能有效提升目标仪器回归预测精度，具有一定的实用性。实验同时也说明迁移学习方法可以在近红外光谱模型传递方面继续深入探索和改进。
5)总结本文的研究结论和创新工作，提出下一步研究工作重点将围绕半监督学习模型的预测输出置信度、异常光谱凸壳判别、特征波长筛选等方面开展研究，逐步建立起基于近红外光谱分析的产品质量评价等应用的技术框架。
In today's era, the science and economy have developed rapidly. The level of theproduction process is becoming more and more automatic and intelligent. Thetraditional means of product quality control can not meet the needs of productdevelopment and control. Near-infrared spectroscopy (NIR) has emerged as a newdetection technology with fast and efficient features. It can greatly improve theefficiency of product quality supervision and management, and now it is widely usedin the oil, medicine, tobacco industry, and etc.
By summarizing the conclusion of early subject named as “Intelligent Methodsof Sensory Evaluation”, it shows that building a sensory evaluation model with goodclassification performance is difficult when the input of the model is chemicalcomposition with incomplete information. The number of components is oftenlimited by detecting with traditional methods of chemical analysis, but near-infraredspectral contains a wealth of component information. The near-infrared spectroscopyalways applied to detect the chemical components of products. In this paper, kinds ofmachine learning methods are applied in the research field of near-infraredspectroscopy. The features of cigarette quality and the components of cigaretteblends is mined from the near-infrared spectroscopy. The high-dimensional spectraldata is directly applied to build the relationships model between product quality andspectra.
Many practices show when NIR is applied to resolve the prediction problem ofcomplex component under the environment with large background noise, thetraditional analysis technology of near-infrared spectroscopy will always need a largeamount of samples, and the model always has poor performance and stability, and it isdifficult to transfer. So the existing modeling technology of NIR should be improved.This paper begins with analyzing the basic principles of modeling of NIR indomesticand aboard research status, semi-supervised learning and transfer learning are introduced for modeling method system of NIR under the inspiration of transductivereference. The paper describes the four key technologies about the high-dimensionaldata processing of NIR, qualitative analysis of spectral, quantitative analysis ofspectral and model transfer. The main content of this paper is as:
1) When the relationship between the observation data and near infrared spectra(NIR) is nonlinear, traditional dimension reduction methods can easily cause to losecharacteristic information, destroy manifold structure and decline the performance ofclassifier and so on. In this paper, a novel algorithm based on semi-supervised kernelneighborhood protection embedding is provided, named as SSKNPE. The kenerldistance is used to transform nonlinear problem into the linear problem in newfeatures space. It can take advantage of the prior classified information from labeleddata, and constraint feature mapping, so the data is mapped from high-dimensionspace to low-dimension space with preserving global structure and local structure.The experimental results show that SSKNPE can effectively improve theclassification ability after dimension reduction. The SSKNPE algorithm is applied tosolve the the cigarette brand identification problem based on the near-infraredspectrum.
2) Because of the inductive reference mechanism, the traditional classifier has theproblem of large prediction risk and training sample number. In this paper, weintroduce transduction reference and semi-supervised learning, and provide a novelsemi-supervised and support vector machines based on affinity propagation clustering(APS4VM). The low density area is found from many large margins by combiningaffinity propagation clustering with chaos optimization. The method can find asupport vector decision surface which can classify samples safely. Though there arefew labeled data in the iris data set and the taste evaluation data set of cigarettes, thegood performance of classifier is obtained based APS4VM algorithm. Sosemi-supervised support vector machine has practical value in engineering application.It is suitable for building the qualitative model of cigarette taste evaluation.
3) The quantitative analysis model based on traditional regression method can notperform well when it meets complex nonlinear problems, particularly when the training samples are not enough. In this paper a novel semi-supervised support vectorregression algorithm (QPSO-LSS3VR) is provided which is based on quantumparticle swarm optimization. The unlabeled samples are estimated by combining theK-nearest neighbor and confidence selection method. The best parameters(γ, λ,σ) ofthe semi-supervised support vector regression model are found by QPSO algorithm.The experiment result of predicting total sugar of cigarette shows that this algorithmcan effectively reduce the standard error of prediction and the cost of modelling whenthere are few training samples, and the algorithm can be applied to build the model ofsugar content prediction of cigarette.
4) The problem is that model versatility between near-infraed spectrometers isbad. The existing model transfer methods are analyzed its non-applicability: thepreparation condition of standard samples for modeling are required harshly, and thepractice operation process is complex, and the model has low prediction performanceafter using traditional statistical model transfer methods.In this paper, a new modeltransfer algorithm of near infrared spectral is innovately provided. It includes the ideaof transfer learning and similarity sample distance metric, named as SM-TrBoostEns.NIR spectral is projected into a low-dimensional space by nonlinear dimensionreduction method. According to metic the similarity between samples, the knowledgetransfers by that useful samples are selected for modeling on the target instrument,and the model is transferred with the combination of transfer boosting technology andensemble learning. The experiments of predicting cigarette total sugar by transferringmodel between two NIR instruments shows that the algorithm can still effectivelyincrease the regression precision under the condition that less standard samples’spectrals are collected on target instrument, so this algorithm has a certain practical.The experiments also shows that transfer learning can be explored and improved indepth for applying to the model transfer of NIR.
5) At last, the research conclusion and innovation is summarized. In the future,more relevant content will be researched, such as the output confidence of semi-supervised learning model, abnormal spectral sample discrimination based onconvex hull, and features wavelength filtering and so on. All results will help us tobuild a new framework of quality evaluation of product based on the near infraredspectra.
引文
[1]陆婉珍主编.现代近红外光谱分析技术(第二版),北京:中国石化出版社,2006
[2]李庆波.近红外光谱分析中若干关键技术的研究[D].天津大学,2002
[3]张建平,谢雯燕,束茹欣等.烟草化学成分的近红外快速定量分析研究[J].烟草科技，1999(3):37-38.
[4]褚小立,袁洪福,陆婉珍等.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4):528-542.
[5]田高友,袁洪福,刘慧颖等.小波变换用于近红外光谱性质分析[J].分析化学,2004,32(9):1125-1130.
[6]王国庆,邵学广.离散小波变换-遗传算法-交互检验法用于近红外光谱数据的高倍压缩与变量筛选[J].分析化学,2005,33(2):191-194
[7]褚小立,袁洪福,王艳斌等.遗传算法用于偏最小二乘方法建模中的变量筛选[J].分析化学,2001,29(4):437-442
[8]陈斌,王豪,林松等.基于相关系数法与遗传算法的啤酒酒精度近红外光谱分析[J].农业工程学报,2005,21(7):99-102.
[9] F.J. Rambla, S. Garrigues, M. de la Guardia, PLS-NIR determination of total sugar, glucose,fructose and sucrose in aqueous solutions of fruit juices, Analytica Chimica Acta,1997,Vol344:41-53
[10] Bochereau L, Beurgine P, et al. A method for prediction by combining data analysis andneural network: application to predictior of apple quality using near infrared spectra
[J].Journal of Agricultural Engineering Research,1992(51):207-216.
[11]瞿海斌,刘晓宣,程翼宇等.中药材三七提取液近红外光谱的支持向量机回归校正方法[J].高等学校化学学报,2004,25(1):39-43.
[12]侯振雨,姚树文,谷永庆等.独立成分分析-支持向量机回归模型及其在近红外光谱分析中的应用[J].河南师范大学学报（自然科学版）,2006,34(2):75-78.
[13]刘旭华,闵顺耕,何雄奎等.近红外光谱逆回归降维定量分析模型[J].光谱学与光谱分析,2011,31(8):2098-2101.
[14]鲍峰伟,彭黔荣,刘景艳等.潜变量聚类分析法在近红外光谱波长范围选择中的应用研究[J].光谱学与光谱分析,2008,28(5):1057-1061.
[15]方利民,林敏.小波聚类方法和近红外光谱技术用于药片种类判别[J].光谱学与光谱分析,2010,30(11):2958-2961.
[16]徐云,吴静珠,王一鸣等.基于近红外光谱的未知类别样品聚类方法[J].农业工程学报,2011,27(8):345-349.
[17]苏谦,邬文锦,王红武等.基于近红外光谱和仿生模式识别玉米品种快速鉴别方法[J].光谱学与光谱分析,2009,29(9):2413-2416.
[18]褚小立,袁洪福,陆婉珍等.光谱多元校正中的模型传递[J].光谱学与光谱分析,2001,21(6):881-885.
[19]褚小立,袁洪福,陆婉珍等.普鲁克分析用于近红外光谱仪的分析模型传递[J].分析化学,2002,30(1):114-119.
[20]王艳斌,袁洪福,陆婉珍等.一种基于目标因子分析的模型传递方法[J].光谱学与光谱分析,2005,25(3):398-401.
[21]黄承伟,戴连奎,董学锋等.结合SNV的分段直接标准化方法在拉曼光谱模型传递中的应用[J].光谱学与光谱分析,2011,31(5):1279-1282.
[22] Williamsion R E, Chaplin J F, Mcclure W F. Near-infrared spectrophotometry of tobaccoleaf for estimating tar yield of smoke [C].40th Tobacco Chem Res Conf, Knoxville,Tenn,1986,40:26-27.
[23] Maha Hana, McClure W F. Applying artificial neural networks..Using near infrared dataⅡ to classify tobacco types and identify native grown tobacco [J]. J Near Infrared Spectros,1997(5):19-25.
[24] Kimy O. The prediction of blending ratio of cut tobacco,expanded stem, and expanded cuttobacco in cigarette using near infrared spectroscopy [J]. J Korean Society of Tob Sci,2000,22(1):76-83.
[25]束茹欣,王国东,张建平等.国产烤烟烟叶的NIRS模式识别[J].烟草科技,2006,(8):12-15,20.
[26]蒋锦锋,李莉,赵明月等.应用近红外检测技术快速测定烟叶主要化学成分[J].中国烟草学报,2006,12(2):8-12.
[27]张勇,丛茜,谢云飞等.烟草组分的近红外光谱和支持向量机分析[J].高等学校化学学报,2009,30(4):697-700.
[28]王家俊,汪帆,马玲等.SIMCA分类法与PLS算法结合近红外光谱应用于卷烟纸的质量控制[J].光谱学与光谱分析,2006,26(10):1858-1862.
[29]周志华,王珏.机器学习及其应用[M].北京:清华大学出版社,2007
[30] F. Rosenblatt. The perceptron: A probabilistic model for information stor-age and organiza-tion in the brain. Psychological Review,1958,65(6):386–408.
[31] J. J. Hopfield. Neural Networks and Physical Systems with Emergent CollectiveComputational Abilities.1982,79:2554-2558.
[32] D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representa-tions byerror propagation. In Parallel distributed processing: explorations in the micro-structure ofcognition, MIT Press,1986,Vol.1:318-362.
[33] J. R. Quinlan.1986. Induction of Decision Trees. Mach. Learn.1986,81-106.
[34] Nearest neighbor pattern classification. by: T. Cover, and P. Hart. In: Information Theory,IEEE Transactions on,1967, Vol.13:21-27.
[35] Vladimir N. Vapnik. The Nature of Statistical Learning Theory. Springer-Verlag New York,Inc., New York, NY, USA.1995.
[36] T. Kohonen, Self-Organizing Map, Springer Verlag,1995
[37] J.B. MacQueen. Some Methods for classification and Analysis of Multivariate Observations,Proceedings of5-th Berkeley Symposium on Mathematical Statistics and Probability,Berkeley, University of California Press,1967,1:281-297
[38] T.G. Dietterich, Machine-Learning Research: Four Current Directions AI Magazine,1997,Vol.18(4):97-135.
[39] Blake C, Merz C J. UCI repository of machine learning databases. University of California,http://www.ics.uci. edu/～mlearn/%MLRepository.html,1998.
[40] Joachims T.Transductive inference for text classification using support vector machine.Inproceedings of the Sixteenth International Conference on Machine Learning[C].MorganKaufmann,1999:148-156.
[41] Sugato Basu, Mikhail Bilenko, and Raymond J. Mooney. A probabilistic framework forsemi-supervised clustering. In Proceedings of the tenth ACM SIGKDD international confe-rence on Knowledge discovery and data mining (KDD '04). ACM, New York,2004:59-68.
[42] Levi Lelis, Jorg Sander,"Semi-supervised Density-Based Clustering," Data Mining,2009Ninth IEEE International Conference on Data Mining,2009:842-847.
[43] Xiaojin Zhu. Semi-Supervised Learning with Graphs. PhD thesis, Carnegie Mellon Univer-sity,2005. CMU-LTI-05-192.
[44] Zhi-Hua Zhou and Ming Li.2005. Tri-Training: Exploiting Unlabeled Data Using ThreeClassifiers. IEEE Trans. on Knowl. and Data Eng.17,11(November2005),1529-1541.
[45]杨剑,王珏,钟宁等.流形上的Laplacian半监督回归[J].计算机研究与发展,2007,44(7):1121-1127.
[46]梁军,陈龙,周卫琪等.基于马尔科夫随机场和鲁棒误差函数的半监督分类研究[J].山东大学学报(理学版),2010,45(11):1-4.
[47]赵莹,张健沛,杨静等.一种改进的分枝定界半监督支持向量机学习算法[J].电子学报,2010,38(2):449-454.
[48]武妍,徐凯.基于增量半监督仿生模式识别的运动想象脑电识别[J].中国生物医学工程学报,2011,30(6):878-884.
[49]唐晓亮,韩敏.一种基于极端学习机的半监督学习方法[J].大连理工大学学报,2010,50(5):771-776.
[50] J. Schmidhuber. On learning how to learn learning strategies. Technical Report FKI-198-94,Fakultat fur Informatik,1994.
[51] R. Caruana. Multitask learning. Machine Learning,28(1):41–75,1997.
[52] Sinno Jialin Pan, James T. Kwok, and Qiang Yang.2008. Transfer learning via dimension-ality reduction. In Proceedings of the23rd national conference on Artificial intelligence(AAAI'08), Anthony Cohn (Ed.), Vol.2. AAAI Press677-682.
[53] Roger Luis, L. Enrique Sucar, and Eduardo F. Morales.2010. Inductive transfer for learningBayesian networks. Mach. Learn.2010.,79:227-255.
[54] Qiang Yang. Transfer Learning beyond Text Classification. In Proceedings of the1st AsianConference on Machine Learning: Advances in Machine Learning (ACML '09),Springer-Verlag, Berlin, Heidelberg,2009,10-22.
[55] Derek Hao Hu, Vincent Wenchen Zheng, and Qiang Yang. Cross-domain activityrecognition via transfer learning. Pervasive Mob. Comput.2011.7(3):344-358.
[56] Wenyuan Dai, Qiang Yang, Gui-Rong Xue and Yong Yu. Boosting for Transfer Learning. InProceedings of the Twenty-Fourth International Conference on Machine Learning (ICML2007), Pages193-200, Corvallis, Oregon, USA, June20-24,2007.
[57]黄育钊.基于样本迁移的多核学习算法研究[D].中山大学,2010.
[58]杨沛,谭琦,丁月华等.一种面向非线性回归的迁移学习模型[J].计算机科学,2009,36(8):212-214,242.
[59]刘晓宣.近红外光谱定性定量技术在中药质量控制中的应用研究[D].浙江大学,2004.
[60]夏俊芳,李小昱,李培武等.小波变换在脐橙维生素C含量近红外光谱预测中的应用[J].中国农业科学,2007,40(8):1760-1766.
[61]邵学广,刘智超,徐恒等.近红外光谱建模中的波长筛选方法研究[C].//全国第二届近红外光谱学术会议论文集.2008:18-23.
[62]栾东磊.近红光谱分析技术在几种水产品中的应用研究[D].中国海洋大学,2009.
[63] Roman Rosipal and Leonard J. Trejo.2002. Kernel partial least squares regression inreproducing kernel hilbert space. J. Mach. Learn. Res.2(March2002),97-123.
[64]李维莉,张亚平.近红外光谱的主成分分析——马氏距离分类法应用于品牌卷烟烟丝的快速鉴别[J].云南农业大学学报,2010,25(2):268-271.
[65]张灵帅,王卫东,谷运红等.近红外光谱的主成分分析-马氏距离聚类判别用于卷烟的真伪鉴别[J].光谱学与光谱分析,2011,31(5):1254-1257.
[66] Alpaydin.E著.机器学习导论[M].北京:机械工业出版社,2009
[67]赵天闻.基于机器学习方法的人脸识别研究[D].上海交通大学,2008.
[68] Tom M. Mitchell,曾华军,张银奎译.机器学习[M].北京:机械工业出版社,2003.
[69]李昱.半监督流形学习算法研究和应用[D].西安电子科技大学,2010.
[70] D.J. Miller, H.S. Uyar, Amixture of experts classifier with learning based on both labelledand unlabelled data, in: Advances in Neural Information Processing Systems (NIPS-9), MITPress, Cambridge, MA,1997, pp.571–577.
[71]杨宁.计算机辅助卷烟配方设计关键技术研究[D].中国海洋大学,2010.
[72]徐蓉,姜峰,姚鸿勋等.流形学习概述[J].智能系统学报,2006,1(1):44-51.
[73]吴晓婷,闫德勤.数据降维方法分析与研究[J].计算机应用研究,2009,26(8):2832-2835.
[74]徐蓉,姜峰,姚鸿勋等.流形学习概述[J].智能系统学报,2006,1(1):44-51.
[75]乔立山.基于图的降维技术研究及应用[D].南京航空航天大学,2009.
[76]戴文渊.基于实例和特征的迁移学习算法研究[D].上海交通大学,2008.
[77]张志敢.知识的应用与迁移[J].长春师范学院学报,2001,20(1):83-85.
[78] Wenyuan Dai, Qiang Yang, Gui-Rong Xue and Yong Yu. Boosting for Transfer Learning. InProceedings of the Twenty-Fourth International Conference on Machine Learning (ICML2007), Pages193-200, Corvallis, Oregon, USA, June20-24,2007
[79] Dai Wenyuan, Xue Guirong, Yang Qiang, et al. Co-clustering based classification forout-of-domain documents[C]//The Thirteenth ACM SIGKDD International Conference onKnowledge Discovery and Data Mining (KDD2007).2007:210-219.
[80] Wenyuan Dai, Yuqiang Chen, Gui-Rong Xue, Qiang Yang, and Yong Yu. TranslatedLearning: Transfer Learning across Different Feature Spaces. Advances in NeuralInformation Processing Systems21(NIPS2008),2008.
[81] Tom M. Mitchell,曾华军,张银奎译.机器学习[M].北京:机械工业出版社,2003.
[82] De Ridder D, Kouropteva O, and Okun O, et al.. Supervised locally linear embedding[C].Artificial Neural Networks and Neural Information Processing-ICANN/ICONIP-2003,Springer,2003:2714,333-341.
[83] SCHOLKOPF B.,SMOLA A. J.,MULLER K. R.Nonlinear component analysis as a kerneleigenvalue problem [J]. Neural Computation,1998,10(5):1299-1319.
[84] Deng Cai, Xiaofei He, and Jiawei Han.Sparse projections over graph. In Proceedings of the23rd national conference on Artificial intelligence(AAAI'08),2008.Vol.2:610-615.
[85] Shuicheng Yan, Dong Xu, Benyu Zhang, and Hong-Jiang Zhang. Graph Embedding: A Gen-eral Framework for Dimensionality Reduction. In Proceedings of the2005IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition (CVPR'05),2005, Vol.2:830-837.
[86]杜卓明,屠宏,耿国华等.KPCA方法过程研究与应用[J].计算机工程与应用,2010,46(7):8-10.
[87]王靖.流形学习的理论与方法研究[D].浙江大学,2006.
[88]洪明坚,温志渝,张小洪等.一种基于流形学习的近红外光谱分析建模方法[J].光谱学与光谱分析,2009,29(7):1793-1796.
[89] Tenenbaum J. B, de Silva V, Langford JC. A global geometric framework for nonlineardimensionality reduction. Science,2000,290(5500):2319-2323.
[90] Roweis S. T, Saul LK. Nonlinear dimensionality reduction by locally linear embedding.Science,2000,290(5500):2323-2326.
[91] X.He，D.Cai，S.Yan，and H．Zhang, Neighborhood Preserving Embedding,ComputerVision,ICCV2005. Tenth IEEE International Conference on.2005, Vol.2:1208-1213.
[92]谷瑞军.基于流形学习的高维空间分类器研究[D].江南大学,2008.
[93] Yang,X.Fu,H.Zha,H.Barlow,J.Semi-supervised nonlinear dimensionality reduction.Machinelearning International Workshop Conference.2006,vol.23:1065-1072
[94]汪炼.基于半监督流形学习的人脸识别算法研究[D].安徽大学,2010.
[95]刘冠群,王庆军,张汝波等.核空间正交及不相关邻域保持鉴别嵌入算法[J].哈尔滨工程大学学报,2011,32(7):938-942
[96] Bennett, K.,&Demiriz, A. Semi-supervised support vector machines. In Advances in NeuralInformation Processing Systems11,1999,368–374.
[97] Wang, J., Shen, X., and Pan, W.(2007). On transductive support vector machines.Contemporary Mathematics.443,7-19.
[98] Y.-F. Li, J. T. Kwok, and Z.-H. Zhou. Semi-supervised learning using label mean. In:Proceedings of the26th International Conference on Machine Learning (ICML'09),2009,633-640.
[99]刘叶青,刘三阳,谷明涛等.一种多项式光滑的半监督支持向量机分类算法[J].计算机科学,2009,36(7):179-181.
[100]王永,程灿,戴明军等.一种半监督支持向量机优化方法[J].工矿自动化,2010,36(12):47-50.
[101] Kennedy J, Eberhart R. Particle swarm optimization. IEEE International Conference onNeural Networks. Perth, Australia.1995.
[102]冯春,谢进,李柏林等.混沌优化算法的研究[J].机械设计与研究,2004,20(z1):304-306.
[103] Frey B J, Dueck D. Clustering by passing messages between data points[J]. Science,2007,315(5814):972-976.
[104] Suykens J.A.K., Lukas L., Van Dooren P., De Moor B., Vandewalle J.,``Least squaressupport vector machine classifiers: a large scale algorithm'', in Proc. of the EuropeanConference on Circuit Theory and Design (ECCTD'99), Stresa, Italy, Sep.1999, pp.839-842
[105]张健沛,赵莹,杨静等.最小二乘支持向量机的半监督学习算法[J].哈尔滨工程大学学报,2008,29(10):1088-1092.
[106]刘解放,陈娜,赵磊等.最小二乘支持机及其数学原理和应用研究[J].河南科技学院学报(自然科学版).2008,36(3):127-129.
[107] Xiaojin Zhu. Semi-Supervised Learning with Graphs. PhD thesis, Carnegie MellonUniversity,2005. CMU-LTI-05-192.
[108] Mikhail Belkin and Partha Niyogi. Laplacian Eigenmaps for dimensionality reduction anddata representation. Neural Computer.2003,15(6):1373-1396.
[109] Z.-H. Zhou and M. Li. Semi-supervised regression with co-training style algorithms. IEEETransactions on Knowledge and Data Engineering,2007,19(11):1479-1493.
[110]周志华.半监督学习中的协同训练风范.见:周志华,王珏主编,机器学习及其应用2007,北京:清华大学出版社,2007,259-275
[111]程玉虎,冀杰,王雪松.基于Help-Training的半监督支持向量回归[J].控制与决策.
[112]盛高斌.基于半监督回归的选择性集成算法及其应用研究[D].浙江工业大学,2009.
[113]马蕾,汪西莉.基于支持向量机协同训练的半监督回归[J].计算机工程与应用,2011,47(3):177-180.
[114]李林,徐硕,安欣等.近红外光谱定量分析的新方法:半监督最小二乘支持向量回归机[J].光谱学与光谱分析,2011,31(10):2702-2705.
[115] Jun Sun, Wenbo Xu, Bin Ye: Quantum-Behaved Particle Swarm Optimization ClusteringAlgorithm. ADMA2006:340-347
[116]张智晟,董存,吴新振.基于量子粒子群优化算法的水电系统经济运行[J].电网技术,2009年,18:68-72.
[117]陈嘉威,周昌乐,张晔晖等.滤光片型近红外仪器模型传递的研究[J].光谱学与光谱分析,2008,28(10):2459-2462.
[118]田高友,褚小立,袁洪福等.小波变换-分段直接校正法用于近红外光谱模型传递研究[J].分析化学,2006,34(7):927-932.
[119]陈斌,王豪.专利算法在白酒酒精度近红外光谱分析模型转移中的应用[J].红外技术,2006,28(4):245-248.
[2]李庆波.近红外光谱分析中若干关键技术的研究[D].天津大学,2002
[3]张建平,谢雯燕,束茹欣等.烟草化学成分的近红外快速定量分析研究[J].烟草科技，1999(3):37-38.
[4]褚小立,袁洪福,陆婉珍等.近红外分析中光谱预处理及波长选择方法进展与应用[J].化学进展,2004,16(4):528-542.
[5]田高友,袁洪福,刘慧颖等.小波变换用于近红外光谱性质分析[J].分析化学,2004,32(9):1125-1130.
[6]王国庆,邵学广.离散小波变换-遗传算法-交互检验法用于近红外光谱数据的高倍压缩与变量筛选[J].分析化学,2005,33(2):191-194
[7]褚小立,袁洪福,王艳斌等.遗传算法用于偏最小二乘方法建模中的变量筛选[J].分析化学,2001,29(4):437-442
[8]陈斌,王豪,林松等.基于相关系数法与遗传算法的啤酒酒精度近红外光谱分析[J].农业工程学报,2005,21(7):99-102.
[9] F.J. Rambla, S. Garrigues, M. de la Guardia, PLS-NIR determination of total sugar, glucose,fructose and sucrose in aqueous solutions of fruit juices, Analytica Chimica Acta,1997,Vol344:41-53
[10] Bochereau L, Beurgine P, et al. A method for prediction by combining data analysis andneural network: application to predictior of apple quality using near infrared spectra
[J].Journal of Agricultural Engineering Research,1992(51):207-216.
[11]瞿海斌,刘晓宣,程翼宇等.中药材三七提取液近红外光谱的支持向量机回归校正方法[J].高等学校化学学报,2004,25(1):39-43.
[12]侯振雨,姚树文,谷永庆等.独立成分分析-支持向量机回归模型及其在近红外光谱分析中的应用[J].河南师范大学学报（自然科学版）,2006,34(2):75-78.
[13]刘旭华,闵顺耕,何雄奎等.近红外光谱逆回归降维定量分析模型[J].光谱学与光谱分析,2011,31(8):2098-2101.
[14]鲍峰伟,彭黔荣,刘景艳等.潜变量聚类分析法在近红外光谱波长范围选择中的应用研究[J].光谱学与光谱分析,2008,28(5):1057-1061.
[15]方利民,林敏.小波聚类方法和近红外光谱技术用于药片种类判别[J].光谱学与光谱分析,2010,30(11):2958-2961.
[16]徐云,吴静珠,王一鸣等.基于近红外光谱的未知类别样品聚类方法[J].农业工程学报,2011,27(8):345-349.
[17]苏谦,邬文锦,王红武等.基于近红外光谱和仿生模式识别玉米品种快速鉴别方法[J].光谱学与光谱分析,2009,29(9):2413-2416.
[18]褚小立,袁洪福,陆婉珍等.光谱多元校正中的模型传递[J].光谱学与光谱分析,2001,21(6):881-885.
[19]褚小立,袁洪福,陆婉珍等.普鲁克分析用于近红外光谱仪的分析模型传递[J].分析化学,2002,30(1):114-119.
[20]王艳斌,袁洪福,陆婉珍等.一种基于目标因子分析的模型传递方法[J].光谱学与光谱分析,2005,25(3):398-401.
[21]黄承伟,戴连奎,董学锋等.结合SNV的分段直接标准化方法在拉曼光谱模型传递中的应用[J].光谱学与光谱分析,2011,31(5):1279-1282.
[22] Williamsion R E, Chaplin J F, Mcclure W F. Near-infrared spectrophotometry of tobaccoleaf for estimating tar yield of smoke [C].40th Tobacco Chem Res Conf, Knoxville,Tenn,1986,40:26-27.
[23] Maha Hana, McClure W F. Applying artificial neural networks..Using near infrared dataⅡ to classify tobacco types and identify native grown tobacco [J]. J Near Infrared Spectros,1997(5):19-25.
[24] Kimy O. The prediction of blending ratio of cut tobacco,expanded stem, and expanded cuttobacco in cigarette using near infrared spectroscopy [J]. J Korean Society of Tob Sci,2000,22(1):76-83.
[25]束茹欣,王国东,张建平等.国产烤烟烟叶的NIRS模式识别[J].烟草科技,2006,(8):12-15,20.
[26]蒋锦锋,李莉,赵明月等.应用近红外检测技术快速测定烟叶主要化学成分[J].中国烟草学报,2006,12(2):8-12.
[27]张勇,丛茜,谢云飞等.烟草组分的近红外光谱和支持向量机分析[J].高等学校化学学报,2009,30(4):697-700.
[28]王家俊,汪帆,马玲等.SIMCA分类法与PLS算法结合近红外光谱应用于卷烟纸的质量控制[J].光谱学与光谱分析,2006,26(10):1858-1862.
[29]周志华,王珏.机器学习及其应用[M].北京:清华大学出版社,2007
[30] F. Rosenblatt. The perceptron: A probabilistic model for information stor-age and organiza-tion in the brain. Psychological Review,1958,65(6):386–408.
[31] J. J. Hopfield. Neural Networks and Physical Systems with Emergent CollectiveComputational Abilities.1982,79:2554-2558.
[32] D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representa-tions byerror propagation. In Parallel distributed processing: explorations in the micro-structure ofcognition, MIT Press,1986,Vol.1:318-362.
[33] J. R. Quinlan.1986. Induction of Decision Trees. Mach. Learn.1986,81-106.
[34] Nearest neighbor pattern classification. by: T. Cover, and P. Hart. In: Information Theory,IEEE Transactions on,1967, Vol.13:21-27.
[35] Vladimir N. Vapnik. The Nature of Statistical Learning Theory. Springer-Verlag New York,Inc., New York, NY, USA.1995.
[36] T. Kohonen, Self-Organizing Map, Springer Verlag,1995
[37] J.B. MacQueen. Some Methods for classification and Analysis of Multivariate Observations,Proceedings of5-th Berkeley Symposium on Mathematical Statistics and Probability,Berkeley, University of California Press,1967,1:281-297
[38] T.G. Dietterich, Machine-Learning Research: Four Current Directions AI Magazine,1997,Vol.18(4):97-135.
[39] Blake C, Merz C J. UCI repository of machine learning databases. University of California,http://www.ics.uci. edu/～mlearn/%MLRepository.html,1998.
[40] Joachims T.Transductive inference for text classification using support vector machine.Inproceedings of the Sixteenth International Conference on Machine Learning[C].MorganKaufmann,1999:148-156.
[41] Sugato Basu, Mikhail Bilenko, and Raymond J. Mooney. A probabilistic framework forsemi-supervised clustering. In Proceedings of the tenth ACM SIGKDD international confe-rence on Knowledge discovery and data mining (KDD '04). ACM, New York,2004:59-68.
[42] Levi Lelis, Jorg Sander,"Semi-supervised Density-Based Clustering," Data Mining,2009Ninth IEEE International Conference on Data Mining,2009:842-847.
[43] Xiaojin Zhu. Semi-Supervised Learning with Graphs. PhD thesis, Carnegie Mellon Univer-sity,2005. CMU-LTI-05-192.
[44] Zhi-Hua Zhou and Ming Li.2005. Tri-Training: Exploiting Unlabeled Data Using ThreeClassifiers. IEEE Trans. on Knowl. and Data Eng.17,11(November2005),1529-1541.
[45]杨剑,王珏,钟宁等.流形上的Laplacian半监督回归[J].计算机研究与发展,2007,44(7):1121-1127.
[46]梁军,陈龙,周卫琪等.基于马尔科夫随机场和鲁棒误差函数的半监督分类研究[J].山东大学学报(理学版),2010,45(11):1-4.
[47]赵莹,张健沛,杨静等.一种改进的分枝定界半监督支持向量机学习算法[J].电子学报,2010,38(2):449-454.
[48]武妍,徐凯.基于增量半监督仿生模式识别的运动想象脑电识别[J].中国生物医学工程学报,2011,30(6):878-884.
[49]唐晓亮,韩敏.一种基于极端学习机的半监督学习方法[J].大连理工大学学报,2010,50(5):771-776.
[50] J. Schmidhuber. On learning how to learn learning strategies. Technical Report FKI-198-94,Fakultat fur Informatik,1994.
[51] R. Caruana. Multitask learning. Machine Learning,28(1):41–75,1997.
[52] Sinno Jialin Pan, James T. Kwok, and Qiang Yang.2008. Transfer learning via dimension-ality reduction. In Proceedings of the23rd national conference on Artificial intelligence(AAAI'08), Anthony Cohn (Ed.), Vol.2. AAAI Press677-682.
[53] Roger Luis, L. Enrique Sucar, and Eduardo F. Morales.2010. Inductive transfer for learningBayesian networks. Mach. Learn.2010.,79:227-255.
[54] Qiang Yang. Transfer Learning beyond Text Classification. In Proceedings of the1st AsianConference on Machine Learning: Advances in Machine Learning (ACML '09),Springer-Verlag, Berlin, Heidelberg,2009,10-22.
[55] Derek Hao Hu, Vincent Wenchen Zheng, and Qiang Yang. Cross-domain activityrecognition via transfer learning. Pervasive Mob. Comput.2011.7(3):344-358.
[56] Wenyuan Dai, Qiang Yang, Gui-Rong Xue and Yong Yu. Boosting for Transfer Learning. InProceedings of the Twenty-Fourth International Conference on Machine Learning (ICML2007), Pages193-200, Corvallis, Oregon, USA, June20-24,2007.
[57]黄育钊.基于样本迁移的多核学习算法研究[D].中山大学,2010.
[58]杨沛,谭琦,丁月华等.一种面向非线性回归的迁移学习模型[J].计算机科学,2009,36(8):212-214,242.
[59]刘晓宣.近红外光谱定性定量技术在中药质量控制中的应用研究[D].浙江大学,2004.
[60]夏俊芳,李小昱,李培武等.小波变换在脐橙维生素C含量近红外光谱预测中的应用[J].中国农业科学,2007,40(8):1760-1766.
[61]邵学广,刘智超,徐恒等.近红外光谱建模中的波长筛选方法研究[C].//全国第二届近红外光谱学术会议论文集.2008:18-23.
[62]栾东磊.近红光谱分析技术在几种水产品中的应用研究[D].中国海洋大学,2009.
[63] Roman Rosipal and Leonard J. Trejo.2002. Kernel partial least squares regression inreproducing kernel hilbert space. J. Mach. Learn. Res.2(March2002),97-123.
[64]李维莉,张亚平.近红外光谱的主成分分析——马氏距离分类法应用于品牌卷烟烟丝的快速鉴别[J].云南农业大学学报,2010,25(2):268-271.
[65]张灵帅,王卫东,谷运红等.近红外光谱的主成分分析-马氏距离聚类判别用于卷烟的真伪鉴别[J].光谱学与光谱分析,2011,31(5):1254-1257.
[66] Alpaydin.E著.机器学习导论[M].北京:机械工业出版社,2009
[67]赵天闻.基于机器学习方法的人脸识别研究[D].上海交通大学,2008.
[68] Tom M. Mitchell,曾华军,张银奎译.机器学习[M].北京:机械工业出版社,2003.
[69]李昱.半监督流形学习算法研究和应用[D].西安电子科技大学,2010.
[70] D.J. Miller, H.S. Uyar, Amixture of experts classifier with learning based on both labelledand unlabelled data, in: Advances in Neural Information Processing Systems (NIPS-9), MITPress, Cambridge, MA,1997, pp.571–577.
[71]杨宁.计算机辅助卷烟配方设计关键技术研究[D].中国海洋大学,2010.
[72]徐蓉,姜峰,姚鸿勋等.流形学习概述[J].智能系统学报,2006,1(1):44-51.
[73]吴晓婷,闫德勤.数据降维方法分析与研究[J].计算机应用研究,2009,26(8):2832-2835.
[74]徐蓉,姜峰,姚鸿勋等.流形学习概述[J].智能系统学报,2006,1(1):44-51.
[75]乔立山.基于图的降维技术研究及应用[D].南京航空航天大学,2009.
[76]戴文渊.基于实例和特征的迁移学习算法研究[D].上海交通大学,2008.
[77]张志敢.知识的应用与迁移[J].长春师范学院学报,2001,20(1):83-85.
[78] Wenyuan Dai, Qiang Yang, Gui-Rong Xue and Yong Yu. Boosting for Transfer Learning. InProceedings of the Twenty-Fourth International Conference on Machine Learning (ICML2007), Pages193-200, Corvallis, Oregon, USA, June20-24,2007
[79] Dai Wenyuan, Xue Guirong, Yang Qiang, et al. Co-clustering based classification forout-of-domain documents[C]//The Thirteenth ACM SIGKDD International Conference onKnowledge Discovery and Data Mining (KDD2007).2007:210-219.
[80] Wenyuan Dai, Yuqiang Chen, Gui-Rong Xue, Qiang Yang, and Yong Yu. TranslatedLearning: Transfer Learning across Different Feature Spaces. Advances in NeuralInformation Processing Systems21(NIPS2008),2008.
[81] Tom M. Mitchell,曾华军,张银奎译.机器学习[M].北京:机械工业出版社,2003.
[82] De Ridder D, Kouropteva O, and Okun O, et al.. Supervised locally linear embedding[C].Artificial Neural Networks and Neural Information Processing-ICANN/ICONIP-2003,Springer,2003:2714,333-341.
[83] SCHOLKOPF B.,SMOLA A. J.,MULLER K. R.Nonlinear component analysis as a kerneleigenvalue problem [J]. Neural Computation,1998,10(5):1299-1319.
[84] Deng Cai, Xiaofei He, and Jiawei Han.Sparse projections over graph. In Proceedings of the23rd national conference on Artificial intelligence(AAAI'08),2008.Vol.2:610-615.
[85] Shuicheng Yan, Dong Xu, Benyu Zhang, and Hong-Jiang Zhang. Graph Embedding: A Gen-eral Framework for Dimensionality Reduction. In Proceedings of the2005IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition (CVPR'05),2005, Vol.2:830-837.
[86]杜卓明,屠宏,耿国华等.KPCA方法过程研究与应用[J].计算机工程与应用,2010,46(7):8-10.
[87]王靖.流形学习的理论与方法研究[D].浙江大学,2006.
[88]洪明坚,温志渝,张小洪等.一种基于流形学习的近红外光谱分析建模方法[J].光谱学与光谱分析,2009,29(7):1793-1796.
[89] Tenenbaum J. B, de Silva V, Langford JC. A global geometric framework for nonlineardimensionality reduction. Science,2000,290(5500):2319-2323.
[90] Roweis S. T, Saul LK. Nonlinear dimensionality reduction by locally linear embedding.Science,2000,290(5500):2323-2326.
[91] X.He，D.Cai，S.Yan，and H．Zhang, Neighborhood Preserving Embedding,ComputerVision,ICCV2005. Tenth IEEE International Conference on.2005, Vol.2:1208-1213.
[92]谷瑞军.基于流形学习的高维空间分类器研究[D].江南大学,2008.
[93] Yang,X.Fu,H.Zha,H.Barlow,J.Semi-supervised nonlinear dimensionality reduction.Machinelearning International Workshop Conference.2006,vol.23:1065-1072
[94]汪炼.基于半监督流形学习的人脸识别算法研究[D].安徽大学,2010.
[95]刘冠群,王庆军,张汝波等.核空间正交及不相关邻域保持鉴别嵌入算法[J].哈尔滨工程大学学报,2011,32(7):938-942
[96] Bennett, K.,&Demiriz, A. Semi-supervised support vector machines. In Advances in NeuralInformation Processing Systems11,1999,368–374.
[97] Wang, J., Shen, X., and Pan, W.(2007). On transductive support vector machines.Contemporary Mathematics.443,7-19.
[98] Y.-F. Li, J. T. Kwok, and Z.-H. Zhou. Semi-supervised learning using label mean. In:Proceedings of the26th International Conference on Machine Learning (ICML'09),2009,633-640.
[99]刘叶青,刘三阳,谷明涛等.一种多项式光滑的半监督支持向量机分类算法[J].计算机科学,2009,36(7):179-181.
[100]王永,程灿,戴明军等.一种半监督支持向量机优化方法[J].工矿自动化,2010,36(12):47-50.
[101] Kennedy J, Eberhart R. Particle swarm optimization. IEEE International Conference onNeural Networks. Perth, Australia.1995.
[102]冯春,谢进,李柏林等.混沌优化算法的研究[J].机械设计与研究,2004,20(z1):304-306.
[103] Frey B J, Dueck D. Clustering by passing messages between data points[J]. Science,2007,315(5814):972-976.
[104] Suykens J.A.K., Lukas L., Van Dooren P., De Moor B., Vandewalle J.,``Least squaressupport vector machine classifiers: a large scale algorithm'', in Proc. of the EuropeanConference on Circuit Theory and Design (ECCTD'99), Stresa, Italy, Sep.1999, pp.839-842
[105]张健沛,赵莹,杨静等.最小二乘支持向量机的半监督学习算法[J].哈尔滨工程大学学报,2008,29(10):1088-1092.
[106]刘解放,陈娜,赵磊等.最小二乘支持机及其数学原理和应用研究[J].河南科技学院学报(自然科学版).2008,36(3):127-129.
[107] Xiaojin Zhu. Semi-Supervised Learning with Graphs. PhD thesis, Carnegie MellonUniversity,2005. CMU-LTI-05-192.
[108] Mikhail Belkin and Partha Niyogi. Laplacian Eigenmaps for dimensionality reduction anddata representation. Neural Computer.2003,15(6):1373-1396.
[109] Z.-H. Zhou and M. Li. Semi-supervised regression with co-training style algorithms. IEEETransactions on Knowledge and Data Engineering,2007,19(11):1479-1493.
[110]周志华.半监督学习中的协同训练风范.见:周志华,王珏主编,机器学习及其应用2007,北京:清华大学出版社,2007,259-275
[111]程玉虎,冀杰,王雪松.基于Help-Training的半监督支持向量回归[J].控制与决策.
[112]盛高斌.基于半监督回归的选择性集成算法及其应用研究[D].浙江工业大学,2009.
[113]马蕾,汪西莉.基于支持向量机协同训练的半监督回归[J].计算机工程与应用,2011,47(3):177-180.
[114]李林,徐硕,安欣等.近红外光谱定量分析的新方法:半监督最小二乘支持向量回归机[J].光谱学与光谱分析,2011,31(10):2702-2705.
[115] Jun Sun, Wenbo Xu, Bin Ye: Quantum-Behaved Particle Swarm Optimization ClusteringAlgorithm. ADMA2006:340-347
[116]张智晟,董存,吴新振.基于量子粒子群优化算法的水电系统经济运行[J].电网技术,2009年,18:68-72.
[117]陈嘉威,周昌乐,张晔晖等.滤光片型近红外仪器模型传递的研究[J].光谱学与光谱分析,2008,28(10):2459-2462.
[118]田高友,褚小立,袁洪福等.小波变换-分段直接校正法用于近红外光谱模型传递研究[J].分析化学,2006,34(7):927-932.
[119]陈斌,王豪.专利算法在白酒酒精度近红外光谱分析模型转移中的应用[J].红外技术,2006,28(4):245-248.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |