An Overview of Principal Component Analysis

Karamizadeh, Sasan; Abdullah, Shahidan M.; Manaf, Azizah Abdul; Zamani, Mazdak; Hooman, Alireza

doi:10.4236/jsip.2013.43b031

Cited by 360 publications

(209 citation statements)

References 11 publications

Supporting

Mentioning

204

Contrasting

Unclassified

Order By: Relevance

“…Also, it increases the efficiency of the model training given the processes are taking place in smaller dimensions. The main disadvantages of PCA are that it is hard to evaluate the covariance matrix with the desired accuracy, and even modest invariance could not be captured by the PCA unless the information is explicitly provided in the training data [25,26]. This method is popular, and it has since been incorporated in many commercial and open source software platforms.…”

Section: Principal Component Analysismentioning

confidence: 99%

Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter

Suleiman

Tight

Quinn

2016

Environ Model Assess

View full text Add to dashboard Cite

This paper examines the application of artificial neural network (ANN) and boosted regression tree (BRT) methods in air quality modelling. The methods were applied to developing air quality models for predicting roadside particle mass concentration (PM 10 , PM 2.5 ) and particle number counts (PNC) based on air pollution, traffic and meteorological data from Marylebone Road in London. Elastic net, Lasso and principal components analysis were used as feature selection methods for the ANN models to reduce the number of predictor variables and improve their generalisation. The performance of the ANN with feature selection (ANN hybrid) and the BRT models was evaluated and compared using statistical performance metrics. The performance parameters include root mean square error (RMSE), fraction of prediction within a factor of two of the observation (FAC2), mean bias (MB), mean gross error (MGE), the coefficient of correlation (R) and coefficient of efficiency (CoE) values. The input variables selected by the elastic net produced the best performing ANN models. The ANN hybrid produced models performed only slightly better than the BRT models. The R values of the ANN elastic net and BRT models were 0.96 and 0.95 for PM 10 , 0.96 and 0.96 for PM 2.5 and 0.89 and 0.87 for PNC, respectively. Their corresponding CoE values were 0.72 and 0.70 for PM 10 , 0.74 and 0.76 for PM 2.5 and 0.81 and 0.71 for PNC respectively. About 80-99% of all the model predictions are within a factor of two of the observed particle concentrations. The BRT models offer more advantages regarding model interpretation and permit feature selection. Therefore, the study recommends the use of BRT over ANN where the model interpretation is a priority.

show abstract

Section: Principal Component Analysismentioning

confidence: 99%

Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter

Suleiman

Tight

Quinn

2016

Environ Model Assess

View full text Add to dashboard Cite

show abstract

“…Briefly, it can be process using principal component analysis many experimental data transposed in matrix form: correlation matrix or covariance matrix. As noted Martin et al (1996), this technique reduce the multidimensional results to smaller dimensions (some principal components), for easier human understanding, without losing the most important information (Henderson, 2006;Karamizadeh et al, 2013). It can be exploited the contribution of each variable to the principal component linear model interpreting the factor loadings (Martin et al, 1996).…”

Section: Introductionmentioning

confidence: 99%

“…Principal components analysis, considered an "algorithm in biometrics" (Karamizadeh et al, 2013), or an "exploratory tool" (Henderson, 2006), is one of the oldest and more utilized multivariate techniques over time. There are many examples that demonstrate the accuracy of information extracted from many experimental data by principal components multivariate analysis facilities: fraud detection in automobile insurance domain (Brockett et al, 2002), digital images classification (Ehsanirad and Kumar, 2010;Ostaszewski et al, 2015), missing data values identification based on probabilistic formula of a theoretical mathematical model (Ilin and Raiko, 2010;Dray and Josse, 2015), pattern classification of drugs in pharmacology (Bober et al, 2011), cancer diagnose (Bair et al, 2006), pattern analysis of wine (Camara et al, 2006;Giaccio and Vicentini, 2008;Fu et al, 2012) or green tea (Fu et al, 2012), animal behavior depending on environmental conditions (Budaev, 2010), quality evaluation of dairy products (Chapman et al, 2000), fruits classification based on qualitative parameters (Zaragoza, 2015), plants diversity (Casas and Ninot, 2003;Henderson, 2006), foliage identification of plant species based on different characteristics (Ehsanirad and Kumar, 2010;Kadir et al, 2012), genetic variability of plants germplasm (Evgenidis et al, 2011;Mahendran et al, 2015), submergence tolerance of flooded plants in river floodplains (Mommer et al, 2006), selection of the most important criteria of Triticum aestivum genotypes to improve genetically the yield of bread wheat (Beheshtizadeh et al, 2013), etc. This multivariate technique use a linear model in orthogonal projection for extractingessential observations based on amount of the data variance (Casas and Ninot, 2003;Henderson, 2006;Giaccio and Vicentini, 2008;Ilin and Raiko, 2010;Karamizadeh et al, 2013;…”

Section: Introductionmentioning

confidence: 99%

PC&CA - Tool for Polyphenols Content Monitoring at the End of Spring Depending on Organic and/or Mineral Fertilization of Perennial Grassland

Hărmănescu¹

2018

UASVMCN-AGR

View full text Add to dashboard Cite

The present research proposed to use Principal Components & Classification Analysis (PC&CA) as tool for the polyphenols content monitoring of perennial forages, at the end of May 2008, from a Romanian grassland fertilized since 2003. The grassland soil was Calcic Luvisol. StatSoft -STATISTICA Version 10 performed the matrix correlation of PC&CA. The ten trials unfertilized or with different fertilization were the PC&CA cases. The active variables of PC&CA were the floristic composition components: Poaceae (GMay08%), Fabaceae (FMay08%), others botanical families group (OMay08%), Calamagrostis epigejos (Ce%), Festuca rupicola (Fr%), Lathyrus nissolia (Ln%), Lathyrus pratensis (Lp%), Filipendula vulgaris (Fv%) and Rosa canina (Rc%). Perennial forages polyphenols content at the end of May 2008 and mineral nitrogen fertilization data were supplementary variables. The correlation coefficient for total polyphenols content (PMay08) and others botanical families group (OMay08%) was positive: 0.704. In addition, the correlation coefficients of PMay08 and Fv% or Rc% were positively and higher than 0.700. Mineral nitrogen application influenced positively the Poaceae component (correlation coefficient -0.915). The PC&CA can represent a helpful tool for polyphenols content monitoring of grassland perennial forages at the end of spring, depending on fertilization.

show abstract

“…However, using only SFA in the evaluation is one-sided (Shi et al, 2010). Principal component analysis (PCA) can be used to accurately determine the weight of each index, and to find a few of the principal components that can control all the variables (Karamizadeh et al, 2013). Xu Jun et al (2007 hold that PCA is a data analysis method by reducing the dimensionality when large multivariate datasets are analyzed.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of drought resistance in Iris germanica L. based on subordination function and principal component analysis

Wei

Qin

et al. 2017

Emir J Food Agric

View full text Add to dashboard Cite

Drought is one of the major environment stresses that have a wide range of impact on plants. In this study, seven physiological indexes including the content of soluble protein (SP), chlorophyll (Chl) and malondialdehyde (MDA), superoxide dismutase (SOD) and peroxidase (POD) activities, leaf relative water content (RWC), rate of water loss (RWL) from excised leaves were measured in leaves of Iris germanica before and after the drought treatment. It was found that the content of MDA and SP, POD and SOD activity increased, while RWL and RWC decreased in response to drought stress. Based on the subordinate function values of seven physiological indexes, seven single indexes were transformed into three principal components namely damage degree, active oxygen removal ability and moisture condition and the composite score (F value) of each iris variety was calculated by principal component analysis (PCA). Based on the F values, 10 iris cultivars could be divided into three groups by cluster analysis (CA): drought-resistance (2 varieties), medium drought-resistance (5 varieties), and low drought-resistance (3 varieties). Meanwhile, optimum regression equation was constructed. Therefore, this work provides a comprehensive and reliable method for evaluating drought resistance in the varieties of Iris germanica.

show abstract

An Overview of Principal Component Analysis

Cited by 360 publications

References 11 publications

Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter

Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter

PC&CA - Tool for Polyphenols Content Monitoring at the End of Spring Depending on Organic and/or Mineral Fertilization of Perennial Grassland

Evaluation of drought resistance in Iris germanica L. based on subordination function and principal component analysis

Contact Info

Product

Resources

About