新的聚类方法综合评价血脂指标

本文结合中心聚类和模糊聚类提出了一个新的聚类方法。该法避免了在迭代过程中出现局部最小问题,并可给出每个样本归入每一类的成分矩阵,因而提供更多的聚类程度的信息。应用这种方法对正常人和冠心病患者的血脂指标进行了综合评价,结果满意。     

基于分层聚类的黑色签字笔笔迹拉曼光谱研究

提出一种结合分层聚类和判别分析对笔迹成分进行分类检验的方法。利用激光显微共聚焦拉曼光谱仪对收集的市面上常见的130支黑色签字笔笔迹样本进行检测。对测量数据进行Savitzky-Golay卷积平滑和Z-score标准化处理,利用组间连接法、组内连接法和离差平方和法三种分层聚类方法对数据进行分类,将三种聚类方法所得分类结果作为判别依据进行判别分析,检验聚类方法的正确率。结合聚类树状图与正确率,最终选择在分类数为4时原始分类结果正确率为100%、留一交叉验证分类结果正确率为98.5%的离差平方和法,提出了适用于黑色签字笔笔迹拉曼光谱数据的分层聚类方法和判别验证方法。     

红外、近红外光谱-簇类的独立软模式方法识别植物调和油脂

用IR,NIR光谱法结合簇类的独立软模式(SIMCA)识别方法对植物油脂进行分类识别,建立了识别二元、三元植物调和油脂的测定方法。应用NIRCal5.2软件的SIMCA技术,分别为所制备的植物调和油脂建立了IR和NIR识别模型,并讨论了光谱处理和数据处理方法来提高模型的分类识别效果。分别以各种植物调和油脂的IR和NIR光谱为变量,随机抽取2/3的样本作训练集,建立了各个调和油的主成分分析(Princi-pal component analysis,PCA)模型;1/3作验证集,对所建模型进行验证识别。用聚类分析-主成分分析(CLU-PCA)方法考察调和油的IR,NIR光谱信息与其纯油的主成分分布。结果显示,在4000～10000cm-1光谱范围内,SIMCA可以对15种二元调和油和2种三元调和油的NIR光谱分别聚类并识别;并对10种二元调和油和2种三元调和油的IR光谱分别聚类并识别。IR以4个波数1099,1119,1746与2855cm-1的吸收值作为分析基础,选择不同的主成分数及数据预处理方法。各种油脂的SIMCA分析的分类精度均为100%,调和油的验证识别准确率100%,最低识别比例为1%,且IR识别灵敏度高于NIR。     

拉曼光谱结合聚类分析法区分检验塑钢窗

应用拉曼光谱法并结合聚类分析对26种不同品牌、厂家、型号的塑钢窗进行了深入的检验研究。采用显微激光拉曼光谱分析技术分别对样品进行检测,得到在780nm激光光源的一阶导数扩展拉曼光谱中,塑钢窗的光谱形态差异显著,荧光背景干扰弱,重叠的谱峰得到有效的分离,可构建具有高鉴别能力的聚类分析模型,采集光谱数据将其定量化,选择离差平方和法作为类间距离,采用欧氏距离作为度量区间表征样品之间的亲疏程度,进行系统聚类分析,同时结合多种方法验证衡量聚类效果,成功将26种塑钢窗样本分为了18类,实现了基于全波段光谱信息结合系统聚类分析建立模型用于准确鉴定塑钢窗种类的目的,为现场物证的区分检验方法提供了一定的理论依据。     

高光谱结合多元统计分析鉴别一次性纸杯样品

为了快速、准确地对现场一次性纸杯物证进行鉴别,提出了一种基于高光谱技术结合PCA、K-Means聚类、Fisher判别分析的识别分类方法。利用高光谱仪对收集的40个不同来源、不同用途的一次性纸杯进行检验,采用主成分分析法对光谱数据进行预处理,从中提取出了11个主成分。借助K-Means算法将40个样品聚为5组,各组分间区分明显。利用Fisher判别分析构建了3组判别函数,经检验函数模型可排除污染客体干扰,分组准确率达100%。     

纹党参与白条党参红外光谱的SIMCA聚类鉴别方法研究

以纹党参和白条党参的红外光谱为聚类分析的对象,研究了红外光谱结合SIMCA聚类分析法对纹党参和白条党参进行识别与分类的可行性.选取400 ～2 000 cm~(-1)范围内的光谱,通过基线补偿(Offset)和散射校正(MSC)等预处理后,采用SIMCA聚类分析法建立识别模型.结果表明,所建模型对纹党参和白条党参的识别率分别达92%和96%,拒绝率均为100%.用盲样对所建模型进行了测试,测试结果全部正确.该法可实现对纹党参和白条党参的快速鉴别.     

UV–化学模式识别不同产地的丹参

建立由UV–化学模式识别法评价丹参质量的方法。分别用正己烷、乙酸乙酯、水、乙醇提取不同产地的丹参,并测绘其紫外光谱。取紫外光谱各波长的吸光度为特征数据,进行主成分分析、聚类分析,对不同产地丹参质量的差异进行了评价。不同溶剂提取液的光谱聚类结果有所差异,可将不同产地丹参聚为3类。UV–化学模式识别技术可以从整体上反应丹参所含成分的差异,可为丹参质量控制与评价提供依据。     

通草类中药微量元素的主成分分析和聚类分析

探讨通草类中药材中微量元素含量与其功效间的相关性。以微量元素含量为指标,运用主成分分析和聚类分析对11种通草类中药的微量元素进行分析。主成分分析结果表明前3个主因子含有通草类中药材微量元素含量84.50%的信息。利用3个主因子模型和聚类分析谱图,解释了11种通草类中药中药的相似性与差异。利用主成分分析和聚类分析法初步得出11种通草类中药的微量元素与其功效存在相关性,为该类中草药的开发利用提供了科学依据和理论基础。     

差分拉曼光谱技术结合K-means聚类法对牙膏的快速分类

建立了差分拉曼光谱技术结合K-means聚类法对牙膏快速分类的方法。对37个牙膏样品编号,将其分别涂抹于载玻片上,晾干,使用差分拉曼光谱仪进行扫描。调用R语言软件中fpc、factoextra、cluster数据库中的na.omit和scale函数对37个牙膏样品的差分拉曼光谱数据进行标准化处理,利用手肘法和Gap Statistic算法优化聚类数。在最佳聚类数为4的条件下,通过K-means聚类法对牙膏样品进行分类,并使用层次聚类分析法进行验证。结果显示,37个牙膏样品被分为4类,并且两种方法的分类结果一致。     

紫外-可见光谱-偏最小二乘法测定渣油四组分含量

用已建立的渣油聚类和识别模型将渣油分为坟渣油、减压渣油和加氢渣油,用紫外－可见光谱－偏最小二乘法（ＵＶ－ＰＬＳ）分别建立了这３种渣油的饱和烃、芳烃、胶质和沥青质四组分测定模型。该方法的测定结果与经典洗脱色谱法（ＥＣ）的结果相吻合。与ＥＣ方法相比,该方法具有快速、操作简单、不需分离,重复性好等特点。     

Towards the identification of plant and animal binders on Australian stone knives

There is limited information regarding the nature of plant and animal residues used as adhesives, fixatives and pigments found on Australian Aboriginal artefacts. This paper reports the use of FTIR in combination with the chemometric tools principal component analysis (PCA) and hierarchical clustering (HC) for the analysis and identification of Australian plant and animal fixatives on Australian stone artefacts. Ten different plant and animal residues were able to be discriminated from each other at a species level by combining FTIR spectroscopy with the chemometric data analysis methods, principal component analysis (PCA) and hierarchical clustering (HC). Application of this method to residues from three broken stone knives from the collections of the South Australian Museum indicated that two of the handles of knives were likely to have contained beeswax as the fixative whilst Spinifex resin was the probable binder on the third.     

Fisher判别在气相色谱-质谱分析助燃剂燃烧残留物中的应用

为寻找一种用于火场助燃剂燃烧残留物鉴定的更为准确、有效的模式识别方法,对7种常见助燃剂在不同载体上的燃烧残留物样品及未知送检样品进行气相色谱-质谱（GC-MS）分析测试,通过特征组分分析鉴定出未知样品中含有汽油成分。同时运用Fisher判别及PCA（主成分分析）/Fisher判别联用两种判别方法对样本数据进行了分析处理,PCA/Fisher判别联用的结果表明送检样本中含有硝基油漆稀料成分,而仅使用Fisher判别的结果表明送检样本中含有93^#汽油。通过将两种分析方法所得结果与GC-MS特征组分分析的结果进行比对发现,Fisher判别能够对7种助燃剂燃烧残留物的样本实现更有效的分类,对未知样本的判别更为有效。该研究结果为火场助燃剂鉴定提供了新的数据分析手段。     

高效液相色谱-质谱法及模式识别法用于鉴别普洱生茶与熟茶

采用高效液相色谱-质谱联用技术及高效液相色谱法对生熟普洱茶中的主要成分进行定性和定量分析。鉴定出普洱茶水溶液中8种主要成分,分别为没食子酸(GA)、没食子酸儿茶素(GC)、表没食子酸儿茶素(EGC)、儿茶素(C)、咖啡因(CAF)、表儿茶素(EC)、表没食子酸儿茶素没食子酸酯(EGCG)和表儿茶素没食子酸酯(ECG)。以这8种成分的含量为指标,对普洱生茶和熟茶各20批进行主成分分析、聚类分析和判别分析,能准确地区分普洱生茶与熟茶。     

Classification of Edible Oils by Infrared Spectroscopy with Optimized K-Means Clustering by a Hybrid Particle Swarm Algorithm

The K-means algorithm has some limitations including dead-unit properties, heavy dependence on the initial choice of cluster centers, convergence to local optima, and sensitivity to the number of clusters. This paper presents an efficient algorithm that optimizes K-means clustering by a hybrid particle swarm algorithm. The modified discrete algorithm is used to select variables and is continuously applied to update cluster centers simultaneously. The nearest center classification is then employed to classify the test samples. The proposed algorithm was applied to discriminate various edible oil varieties by employing Fourier transform infrared spectroscopy. As a comparison, the common K-means clustering, principal component analysis, and partial least squares techniques were also applied to classify these edible oil samples. Results demonstrated that the proposed method is an accurate and rapid strategy for identifying edible oils.     

Classification of gasoline data obtained by gas chromatography using a piecewise alignment algorithm combined with feature selection and principal component analysis

A fast and objective chemometric classification method is developed and applied to the analysis of gas chromatography (GC) data from five commercial gasoline samples. The gasoline samples serve as model mixtures, whereas the focus is on the development and demonstration of the classification method. The method is based on objective retention time alignment (referred to as piecewise alignment) coupled with analysis of variance (ANOVA) feature selection prior to classification by principal component analysis (PCA) using optimal parameters. The degree-of-class-separation is used as a metric to objectively optimize the alignment and feature selection parameters using a suitable training set thereby reducing user subjectivity, as well as to indicate the success of the PCA clustering and classification. The degree-of-class-separation is calculated using Euclidean distances between the PCA scores of a subset of the replicate runs from two of the five fuel types, i.e., the training set. The unaligned training set that was directly submitted to PCA had a low degree-of-class-separation (0.4), and the PCA scores plot for the raw training set combined with the raw test set failed to correctly cluster the five sample types. After submitting the training set to piecewise alignment, the degree-of-class-separation increased (1.2), but when the same alignment parameters were applied to the training set combined with the test set, the scores plot clustering still did not yield five distinct groups. Applying feature selection to the unaligned training set increased the degree-of-class-separation (4.8), but chemical variations were still obscured by retention time variation and when the same feature selection conditions were used for the training set combined with the test set, only one of the five fuels was clustered correctly. However, piecewise alignment coupled with feature selection yielded a reasonably optimal degree-of-class-separation for the training set (9.2), and when the same alignment and ANOVA parameters were applied to the training set combined with the test set, the PCA scores plot correctly classified the gasoline fingerprints into five distinct clusters.     

交叉响应传感器鉴别β受体激动剂及其混合体系

开发了一种鉴别β受体激动剂的新型阵列传感器。该传感器由8种传感物质构成,使用96孔板酶标仪采集响应数据,结合主成分分析(PCA)、分层聚类分析(HCA)、判别分析(LDA)等模式识别方法进行数据处理,对5类β受体激动剂及其混合物进行检测。PCA结果表明,该传感器主要是基于空间结构以及氢键作用实现对β受体激动剂的识别;HCA结果显示,93个分析样本归类正确;LDA结果显示,该传感器对于β受体激动剂识别的准确率达98.9%。本方法在β受体激动剂的检测中有潜在应用价值。     

Development of colorimetric sensor array for discrimination of herbal medicine

Today, traditional systems of medicines (such as herbal distillates) become important resources for providing healthcare benefits. The ability to discriminate among closely similar herbal products is critical to ensure their efficacy. This article proposes a pattern-based recognition approach for the rapid discrimination of herbal distillates using a low-cost and sensitive colorimetric sensor array composed of 25 indicators. The color changes of the sensor exposed to the vapor of the herbal distillates can be monitored easily with an ordinary flatbed scanner. The digital representation of the array response was analyzed with hierarchical clustering analysis (HCA) and principal component analysis (PCA). Using new variable selection strategy, 6 indicators among the 25 employed indicators were selected as discriminant elements of the array. So, a complete discrimination (with 100% accuracy) of 46 herbal distillates was achieved. The proposed sensor represented a better resolution when analytes were placed in an oven at 85 °C for 45 min. This colorimetric sensor array demonstrates excellent potential for quality assurance/control applications of herbal distillates.     

Fingerprint Chromatogram Analysis of Radix Glehniae by LC Coupled with Hierarchical Clustering Analysis

For the first time a validated liquid chromatography method coupled with hierarchical clustering analysis has been developed for the study of fingerprint chromatograms of extracts from the Radix Glehniae of Glehnia littorali. Liquid chromatography with gradient elution was performed on extracts from 23 plant samples collected from different geographical locations. Ten of twenty-three plant samples were selected using hierarchical clustering analysis and employed to establish the fingerprint. The fingerprint was established and ten chromatographic peaks were selected as characteristic peaks and panaxynol was used as reference standard compounds. The fingerprint chromatograms had a good stability, precision, and reproducibility. This method is a very reliable and useful method for assessment of the quality of Radix Glehniae.     

基于1H-NMR的模式识别方法应用于异常黑胆质糖尿病患者的尿液代谢组学研究

探讨核磁共振氢谱结合模式识别方法应用于异常黑胆质糖尿病患者的尿液代谢组研究可行性。对32 例异常黑胆质糖尿病患者和29 例健康人尿液进行核磁共振氢谱检测,采用主成分分析(principal component analysis, PCA)、偏最小二乘法判别分析(partial least squares dis-criminant analysis, PLS-DA)、正交偏最小二乘法判别分析(orthogonal to partial least squares discriminant analysis,OPLS-DA)进行模式识别分析,比较3种模式识别方法的判别能力。运用3种模式识别均可以对2组数据进行有效的区分,但OPLS-DA较PCA、P[1]LS-DA更加有效,不仅提高了模式识别方法的判断能力,可以清楚的判断两组中有差异的代谢物。基于核磁共振氢谱结合模式识别分析方法可以为异常黑胆质糖尿病代谢标志物的寻找提供理论依据。OPLS-DA的模式识别方法较其它2种方法更具优势,在揭示维医理论本质上有着广阔的应用前景。