Use Of Principal Component Analysis Research Articles

Diversity in decline: The changing suburban fortunes of Tokyo Metropolis

The suburban territory of Tokyo Metropolis experiences new path-dependent, multifaceted shrinking processes, especially in outer reaches, which are considered a product of synchronized changes in socio-demographic, economic, and political and administrative (fiscal) dimensions. Outer suburban shrinkage is an urgent policy and planning issue for Tokyo's suburban sustainability while Tokyo itself emerges as something of a frontier case globally. We analyze different growth/decline transformation types of Tokyo's suburban municipalities in both pre-bubble growth and post-bubble shrinkage periods with the combined use of Principal Component Analysis and Cluster Analysis, and identify their different trajectories in the contextual transition from growth to shrinkage. By making direct comparisons between growth and shrinkage from a long-term evolutionary perspective, we identify the following: 1) outer suburban impoverishment, 2) different processes of suburban shrinkage from suburban growth, and 3) outer suburban divergence. Suburban shrinkage is revealed as more complicated than suburban growth, generating equally complex challenges for policy responses that will soon reverberate globally. In conclusion, we underline the importance for policy makers worldwide to understand the specificities of outer suburban shrinkage even during periods of suburban growth in order to ensure suburban resilience across different phases of urbanization.

Determination of the Volatile Compounds in Five Varieties of Piper betle L. from Bangladesh Using Simultaneous Distillation Extraction and Gas Chromatography/Mass Spectrometry (SDE-GC/MS)

Five varieties of Piper betle L. var. leaves including Bangla, Sanchi, Misti, Khasia, and BARI Paan 3 from Bangladesh were subjected to the analysis of their volatile compounds. Simultaneous distillation extraction (SDE) with 1:1 (v/v) n-pentane:diethylether was used for the isolation of the volatile compounds while the analysis was done by gas chromatography/mass spectrometry (GC/MS). From the results, the Misti betel leaf variety showed the highest concentration of volatile compounds (13959 mg/kg) followed by BARI Paan 3 (11684 mg/kg), Khasia (11110 mg/kg), Sanchi (6959 mg/kg) and Bangla (4347 mg/kg). A total of 101 compounds were identified in which 42 were present in all of the varieties but at varying concentrations. The present study reports 50 volatile compounds in betel leaves for the first time in the literature. Eugenol was found in all of the varieties at the highest concentration value followed by β-caryphyllene, γ-muurolene, valencene, eucalyptol, chavicol and caryphellene oxide. Based on the volatile compounds, the five varieties of betel leaf were differentiated into three clusters by the use of principal component analysis (PCA). Hierarchical cluster analysis (HCA) also divided the leaf varieties into three clusters composed of Misti and BARI Paan 3, Bangla and Sanchi, and Khasia.

Combined use of principal component analysis and artificial neural network approach to improve estimates of PM2.5 personal exposure: A case study on older adults

Accurate exposure estimate of the air pollutant PM2.5 is required to evaluate its health impacts in epidemiological studies, due to its adverse effects on human's respiratory and cardiovascular systems. However, traditional personal sampling is time and cost consuming. Thus, modeling techniques are needed to accurately predict the personal exposure level to PM2.5. In this study, a total of 117 older adults over 60 were recruited in Tianjin, a heavily polluted city in northern China, for indoor, outdoor and personal PM2.5 sampling. Eighteen variables which may increase the exposure level of older adults were recorded for artificial neural network (ANN) simulation. Four modeling techniques, including time-integrated activity modeling, Monte Carlo simulation, ANN modeling, and combined use of principal component analysis (PCA) and ANN model, were used to evaluate their ability for predicting real exposure values of PM2.5. The results of traditional time-weighted activity modeling showed the lowest correlation with measured values with R2 of 0.57 and 0.42 in winter and summer, respectively. For Monte Carlo simulation, high correlation was obtained (R2 of 0.93 and 0.92 in winter and summer, respectively) between percentiles of the predicted and the real exposure values. Compared with the simple ANN models, the combined use of PCA and ANN produced the most accurate results with R2 of 0.99 and RMSE lower than 15. Since the information of the input variables for the PCA-ANN model can be obtained from the questionnaire and fixed air quality monitoring sites, this technique shows a great potential in predicting personal exposure level to the air pollutant because no additional concentration measurement is needed.

Animal welfare and gender: a nexus in awareness and preference when choosing fresh beef meat?

The modern consumer is now more attentive towards animal welfare practices and this represents an important factor when purchasing meat, whereby ethical, sociological and economic implications are evaluated. In addition, the socio-demographic characteristics of consumers evidence different sensitivities with regard to selection patterns and consumption styles. This study aims to explore the role of Gender in beef meat purchasing preferences, assessing consumer awareness of responsibility towards animal welfare, through the use of cross-tabulation with χ2 to test the different behaviour of men and women and the use of principal component analysis and cluster analysis to classify attitudes of choice according to gender. Among the research aims, this study examined consumer attitudes towards certain ‘ethically incorrect’ animal products, as well as their awareness of the institutional responsibility in controlling animal welfare standards during the meat production process. The study conducted in Northwest Italy, involving 512 respondents, shows that women are more sensitive to AW aspects and place trust in those responsible for certification of animal welfare standards, such as veterinarians and consumer associations, and also shows that it is possible to identify an ‘animal welfare sensitive’ profile of meat consumer. HIGHLIGHTS Modern consumer evaluates ethical, sociological and economic implications in animal friendly meat purchasing process Gender affects awareness of the responsibilities of veterinary, public health control bodies and consumer associations to verify animal welfare Cluster highlighted consumer differences in perception towards animal welfare

Concentrations, Sources and Human Health Risk Assessment of 12 Dioxin-Like Polychlorinated Biphenyls (DL-PCBs) in the Urban Soils of Baraki, Algiers (Algeria) with Determination by Gas Chromatography - Tandem Mass Spectrometry (GC-MS/MS)

The concentrations of twelve dioxin-like polychlorobiphenyls (DL-PCB) were determined in fifty-seven soils by gas chromatography coupled to tandem mass spectrometry (GC-MS/MS). The investigated soils were collected in a 72 km2 area close to Algiers, the Algerian capital. The highest cumulated DL-PCB concentration was close to 82 µg kg−1. The use of principal component analysis (PCA) ensured the tentative identification of the contamination sources, which were found to be associated with the release into the environment of PCB commercial mixtures from electrical transformers or from open burning of household wastes. The associated toxic equivalent quantity (TEQ) relative to the 12 DL-PCBs did not exceeded 2.8 ng.kg−1. The use of the United States Environmental Protection Agency guidelines ensured to define that neither the cancer potential human health risk nor the non-carcinogenic risk were significantly increased by the DL-PCB concentrations in the investigated soils.

Flow profile characteristics in Fontan circulation are associated with the single ventricle dilation and function: principal component analysis study.

The Fontan circulation is characterized as a nonpulsatile flow propagation without a pressure-generating ventricle. However, flow through the Fontan circulation still exhibits oscillatory waves as a result of pressure changes generated by the systemic single ventricle. Identification of discrete flow patterns through the Fontan circuit may be important to understand single ventricle performance. Ninety-seven patients with Fontan circulation underwent phase-contrast MRI of the right pulmonary artery, yielding subject-specific flow waveforms. Principal component (PC) analysis was performed on preprocessed flow waveforms. Principal components were then correlated with standard MRI indices of function, volume, and aortopulmonary collateral flow. The first principal component (PC) described systolic versus diastolic-dominant flow through the Fontan circulation, accounting for 31.3% of the variance in all waveforms. The first PC correlated with end-diastolic volume (R = 0.34, P = 0.001), and end-systolic volume (R = 0.30, P = 0.003), cardiac index (R = 0.51, P < 0.001), and the amount of aortopulmonary collateral flow (R = 0.25, P = 0.027)-lower ventricular volumes and a smaller volume of collateral flow-were associated with diastolic-dominant cavopulmonary flow. The second PC accounted for 19.5% of variance and described late diastolic acceleration versus deceleration and correlated with ejection fraction-diastolic deceleration was associated with higher ejection fraction. Principal components describing the diastolic flow variations in pulmonary arteries are related to the single ventricle function and volumes. Particularly, diastolic-dominant flow without late acceleration appears to be related to preserved ventricular volume and function, respectively.NEW & NOTEWORTHY The exact physiological significance of flow oscillations of phasic and temporal flow variations in Fontan circulation is unknown. With the use of principal component analysis, we discovered that flow variations in the right pulmonary artery of Fontan patients are related to the single ventricle function and volumes. Particularly, diastolic-dominant flow without late acceleration appears to be related to more ideal ventricular volume and systolic function, respectively.

Critical Success Strategies for Competitive Advantage of Indigenous Construction Firms in Developing Countries: A Ghana Study

Though a plethora of studies have suggested indigenous construction firms in developing countries are competitively disadvantaged and risk being marginalized in the competition posed by their foreign counterparts, there is lack of empirical study that identifies the critical success strategies for competitive advantage of indigenous construction firms in developing countries to adduce empirical findings towards policy direction and formulation. Thus, this study identifies the critical success strategies for competitive advantage of indigenous construction firms in developing countries using Ghana as a case study and suggests strategies for competitive advantage of indigenous construction firms. The use of principal component analysis and varimax rotation in data analysis aided in reducing the large set of success strategies to four principal components. Using a questionnaire survey, 667 respondents were invited to rate 21 success strategies identified from literature based on their knowledge and experience. Correlations between the 21 variables showed that four key components underlay the critical success strategies for competitive advantage of indigenous construction firms in developing countries, namely tendering, contract, client-centred and branding strategies. Findings and recommendations of this study may be useful to construction stakeholders who are seeking innovative ways to enhance indigenous construction firms’ competitive advantage and policy direction.

Feature selection using principal component analysis and genetic algorithm

Feature engineering is the way toward utilizing domain knowledge of the records to build features that in turn assist Machine Learning (ML) algorithms to provide efficient results. It is crucial to the utilization of ML and is both difficult and costly. The next buzz word after big data is feature engineering, which involves both feature selection and feature extraction. Feature Selection (FS also called attribute selection) is a procedure of selecting a subset of pertinent features for use in model building. It is an optimization problem. In our case, we have used principal component analysis for feature transformation followed by genetic algorithm to select optimal feature set and in the last, decision tree as a classifier. The proposed approach shows that use of principal component analysis before genetic algorithms improves the accuracy of the model with less number of features.

Advancing characterisation with statistics from correlative electron diffraction and X-ray spectroscopy, in the scanning electron microscope

The routine and unique determination of minor phases in microstructures is critical to materials science. In metallurgy alone, applications include alloy and process development and the understanding of degradation in service. We develop a correlative method, exploring superalloy microstructures, which are examined in the scanning electron microscope (SEM) using simultaneous energy dispersive X-ray spectroscopy (EDS) and electron backscatter diffraction (EBSD). This is performed at an appropriate length scale for characterisation of carbide phases’ shape, size, location, and distribution. EDS and EBSD data are generated using two different physical processes, but each provide a signature of the material interacting with the incoming electron beam. Recent advances in post-processing, driven by ‘big data’ approaches, include use of principal component analysis (PCA). Components are subsequently characterised to assign labels to a mapped region. To provide physically meaningful signals, the principal components may be rotated to control the distribution of variance. In this work, we develop this method further through a weighted PCA approach. We use the EDS and EBSD signals concurrently, thereby labelling each region using both EDS (chemistry) and EBSD (crystal structure) information. This provides a new method of amplifying signal-to-noise for very small phases in mapped regions, especially where the EDS or EBSD signal is not unique enough alone for classification.

Morphometrics of China's Loess Plateau: The spatial legacy of tectonics, climate, and loess deposition history

The Loess Plateau in China is the largest and deepest loess deposit in the world. The surficial geology is dominated by Quaternary loess deposits of varying thicknesses, which exhibit some of the fastest erosion rates in the world. Although the plateau has been traditionally thought of as morphologically homogenous, river network morphometrics in this area are spatially variable, suggesting that landscape evolution processes across the plateau are sensitive to variability in local conditions. We divided the plateau into sub-basins and examined six DEM-derived morphometric parameters: Aspect Ratio, Anisotropic Coefficient Variation (ACV), Bifurcation Ratio, Concavity, Steepness, and Mean Junction Angle. The sub-basins exhibited low aspect ratios, high bifurcation ratios and large junction angles relative to landscapes around the world, however they varied intra-regionally relative to each other. Through the use of Principal Component Analysis and Cluster Analysis, we observed specific spatial variation in morphometrics attributable to differences in climate, lithology and tectonics along a NW-SE transect through the plateau. Sub-basins in the elevated plateau regions are more oval with higher aspect ratios and lower ACV values while those located in down-dropped valley regions have smaller junction angles, lower concavity and steepness index. Sub-basins in the transition area between down-dropped and elevated parts of the plateau are distinguished by their high bifurcation ratios and elongated shape. We propose a conceptual model to explain the relationship between these morphometrics and the local environmental conditions.

Study of Climatic Factors Affecting the Productivity of Cotton and its Instability

Cotton is an important commercial crop in India. The present study focuses on measurement of variability pattern of cotton yield and use of principal component analysis for developing cotton yield forecast model for Hisar district of Haryana (India). Instability index has been observed to study the variability behavior of cotton yield in the district. Time series data on cotton yield and fortnightly data of five weather variables for the crop season for 38 years (1980-91 to 2017-18) have been used. In all, three models have been developed by using direct weather variables, PC scores and components with higher loading as regressors and developed models have been used to forecast yield for four subsequent years 2014-15 to 2017-18 (which were not included in model development). The model with PC scores was found to be most appropriate to provide reliable yield forecast.

Mapping genomic regions affecting milk traits in Sarda sheep by using the OvineSNP50 Beadchip and principal components to perform combined linkage and linkage disequilibrium analysis

BackgroundThe detection of regions that affect quantitative traits (QTL), to implement selection assisted by molecular information, remains of particular interest in dairy sheep for which genetic gain is constrained by the high costs of large-scale phenotype and pedigree recording. QTL detection based on the combination of linkage disequilibrium and linkage analysis (LDLA) is the most suitable approach in family-structured populations. The main issue in performing LDLA mapping is the handling of the identity-by-descent (IBD) probability matrix. Here, we propose the use of principal component analysis (PCA) to perform LDLA mapping for milk traits in Sarda dairy sheep.MethodsA resource population of 3731 ewes belonging to 161 sire families and genotyped with the OvineSNP50 Beadchip was used to map genomic regions that affect five milk traits. The paternally and maternally inherited gametes of genotyped individuals were reconstructed and IBD probabilities between them were defined both at each SNP position and at the genome level. A QTL detection model fitting fixed effects of principal components that summarize IBD probabilities was tested at each SNP position. Genome-wide (GW) significance thresholds were determined by within-trait permutations.ResultsPCA resulted in substantial dimensionality reduction, in fact 137 and 32 (on average) principal components were able to capture 99% of the IBD variation at the locus and genome levels, respectively. Overall, 2563 positions exceeded the 0.05 GW significance threshold for at least one trait, which clustered into 75 QTL regions most of which affected more than one trait. The strongest signal was obtained for protein content on Ovis aries (OAR) chromosome 6 and overlapped with the region that harbours the casein gene cluster. Additional interesting positions were identified on OAR4 for fat content and on OAR11 for the three yield traits.ConclusionsPCA is a good strategy to summarize IBD probabilities. A large number of regions associated to milk traits were identified. The outputs provided by the proposed method are useful for the selection of candidate genes, which need to be further investigated to identify causative mutations or markers in strong LD with them for application in selection programs assisted by molecular information.

Improved fetal blood oxygenation and placental estimated measurements of diffusion-weighted MRI using data-driven Bayesian modeling.

PurposeMotion correction in placental DW‐MRI is challenging due to maternal breathing motion, maternal movements, and rapid intensity changes. Parameter estimates are usually obtained using least‐squares methods for voxel‐wise fitting; however, they typically give noisy estimates due to low signal‐to‐noise ratio. We introduce a model‐driven registration (MDR) technique which incorporates a placenta‐specific signal model into the registration process, and we present a Bayesian approach for Diffusion‐rElaxation Combined Imaging for Detailed placental Evaluation model to obtain individual and population trends in estimated parameters.MethodsMDR exploits the fact that a placenta signal model is available and thus we incorporate it into the registration to generate a series of target images. The proposed registration method is compared to a pre‐existing method used for DCE‐MRI data making use of principal components analysis. The Bayesian shrinkage prior (BSP) method has no user‐defined parameters and therefore measures of parameter variation in a region of interest are determined by the data alone. The MDR method and the Bayesian approach were evaluated on 10 control 4D DW‐MRI singleton placental data.ResultsMDR method improves the alignment of placenta data compared to the pre‐existing method. It also shows a further reduction of the residual error between the data and the fit. BSP approach showed higher precision leading to more clearly apparent spatial features in the parameter maps. Placental fetal oxygen saturation (FO2) showed a negative linear correlation with gestational age.ConclusionsThe proposed pipeline provides a robust framework for registering DW‐MRI data and analyzing longitudinal changes of placental function.

Human Facial Emotion Recognition using Adaptive Sigmoidal Transfer Function in MLP Neural Network

The human face is very sensitive towards inner feelings particularly with different state of mind under various conditions. The facial expression has used in computer vision to understand the human response against stimuli. But the facial expression is also having the nature of variability and controllability hence its complete generalization from a computer vision point of view is very difficult and challenging, though acceptable performances can be achieved. In this paper, a two stage based facial expression recognition model which carry the Principal component analysis as a feature extractor in the first stage and self-adaptive based activation function in feedforward neural network as a classifier in the second stage have applied. Use of principal component analysis reduces the dimension of features while the adaptive slope of transfer function provides another parameter along with weights to change in making learning faster and accurate. Six most dominant state of facial emotion like angry, surprise, sadness, normal, happy and fear have considered in this paper and performances have been tested over variable expressions. The benefit of the proposed model of self-adaptive activation function has verified through the benchmark XOR problem classification.

Analysis of Particulate Matter Concentration Variability and Origin in Selected Urban Areas in Poland

The work presents the results of research and analyses related to measurements of concentration and chemical composition of three size fractions of particulate matter (PM), PM10, PM2.5 and PM1.0. The studies were conducted in the years 2014–2016 during both the heating and non-heating season in two Polish cities: Wrocław and Poznań. The studies indicate that in Wrocław and Poznań, the highest annual concentrations of particulate matter (PM1.0, PM2.5, and PM10) were observed in 2016, and the mean concentrations were respectively equal to 18.16 μg/m3, 30.88 μg/m3 and 41.08 μg/m3 (Wrocław) and 8.5 μg/m3, 30.8 μg/m3 and 32.9 μg/m3 (Poznań). Conducted analyses of the chemical composition of the particulate matter also indicated higher concentrations of organic and elemental carbon (OC and EC), and water-soluble ions in a measurement series which took place in the heating season were studied. Analyses with the use of principal component analysis (PCA) indicated a dominating percentage of fuel combustion processes as sources of particulate matter emission in the areas considered in this research. Acquired results from these analyses may indicate the influence of secondary aerosols on air quality. In the summer season, a significant role could be also played by an influx of pollutants—mineral dust—originating from outside the analyzed areas or from the resuspension of mineral and soil dust.

Impact identification using nonlinear dimensionality reduction and supervised learning

Real-time monitoring systems that can automatically locate and identify impacts as they occur have become increasingly attractive for ensuring safety and preventing catastrophic accidents in aerospace structures. In most cases, a set of piezoelectric transducers distributed over the structure captures strain–time data, which are preprocessed to extract relevant features that are fed to a supervised learning algorithm to detect, locate, and quantify impacts. The best results achieved to date in feature extraction for impact identification have been obtained with the use of principal component analysis (PCA). However, this technique cannot handle complex nonlinear data. The primary contribution of this study is the implementation of a novel impact identification algorithm that uses a supervised learning algorithm called linear approximation with maximum entropy (LME) in conjunction with different linear and nonlinear dimensionality reduction techniques, including PCA, kernel PCA, Isomap, local linear embedding (LLE), and multilayer autoencoders. The performance of LME with the different reduction techniques is tested with two experimental applications. The results show that the techniques that do not employ graphs, such as PCA, kernel PCA, and autoencoders, perform better, and the method that provides the best results is LME in conjunction with autoencoders. It is further demonstrated that LME with autoencoders works better than the algorithms available in the literature for similar problems.

Variation Prognostic Model on Port Infrastructures: Employment of Principal Component Analysis for Factor Extraction Instigating Disparity on Planning and Implementation Phase

Port is a transportation and communication marine infrastructure that is established for ship berthing and dockage, loading and unloading of passenger, storage of goods and other operational agendas that would impact a region or country in terms of social and economic aspects. Variation order is the modification on the design whilst construction is on-going. This is inevitable in a project. Through the utilization of Principal Component Analysis with a Cronbach’s Alpha of 0.930 and Kaiser-Meyer-Olkin Value of 0.763, nine components weighted, these are Actual Concreting and Precasting on-site for Structure formation, Meteorological Conditioning affecting Weather and Tides, Test Pile Apportionment on Both design and construction, Off-shore soil in-situ Properties, Behaviour and Suitability on fills, Inaccuracy of Geotechnical, Topographical and Hydrographical surveys, Project Release and Initiation Timing on engineering departments, Sudden unforeseen Turbulences during Pile Driving, Coordination of Planning and Implementation Departments, and Unavailability of Material Specification to Actual site from design. After extracting the regressor, assessment of completed port projects with variation and multiple linear regression was then employed to generate the model. The model had an R2 of 0.777 and an R of 0.882. Finally, the model was evaluated and has shown a significant value. Hence, a satisfactory breakthrough in an attempt to develop a model regarding variation order on port projects was established.

Visualizing the Complexity of the Athlete-Monitoring Cycle Through Principal-Component Analysis.

To discuss the use of principal-component analysis (PCA) as a dimension-reduction and visualization tool to assist in decision making and communication when analyzing complex multivariate data sets associated with the training of athletes. Using PCA, it is possible to transform a data matrix into a set of orthogonal composite variables called principal components (PCs), with each PC being a linear weighted combination of the observed variables and with all PCs uncorrelated to each other. The benefit of transforming the data using PCA is that the first few PCs generally capture the majority of the information (ie,variance) contained in the observed data, with the first PC accounting for the highest amount of variance and each subsequent PC capturing less of the total information. Consequently, through PCA, it is possible to visualize complex data sets containing multiple variables on simple 2D scatterplots without any great loss of information, thereby making it much easier to convey complex information to coaches. In the future, athlete-monitoring companies should integrate PCA into their client packages to better support practitioners trying to overcome the challenges associated with multivariate data analysis and interpretation. In the interim, the authors present here an overview of PCA and associated R code to assist practitioners working in the field to integrate PCA into their athlete-monitoring process.

Assessing and solving multicollinearity in sediment transport prediction models using principal component analysis

ABSTRACT Predicting sediment transport in a river system is necessary when attempting to monitor river changes due to incoming particle deposition from upstream reaches. Previous studies have used various techniques such as probabilistic, regression, and equal mobility approaches to make predictions. However, issues associated with multicollinearity occur when using the regression approach. Significant correlations between the explanatory variables in a regression equation can increase the coefficient variance estimates, which make these estimates very sensitive to minor changes. Thus, we propose the use of principal component analysis as a solution to these issues. We observed that the variance inflation factor is less than 10, which indicates the nonexistence of issues associated with multicollinearity. Fitting components to the regression equation create a new prediction model for sediment transport, which improves upon previous equations using the same technique.

Diagnosis of a battery energy storage system based on principal component analysis

This paper proposes the use of principal component analysis (PCA) for the state of health (SOH) diagnosis of a battery energy storage system (BESS) that is operating in a renewable energy laboratory located in Chocó, Colombia. The presented methodology allows the detection of false alarms during the operation of the BESS. The principal component analysis model is applied to a parameter set associated to the capacity, internal resistance and open circuit voltage of a battery energy storage system. The parameters are identified from experimental data collected daily. The PCA model retains the first 5 components that collect 80.25% of the total variability. During the test under real operation contidions, PCA diagnosed a degradation of state of health fastest than the comercial battery controller. A change in the charging modes lead to a battery recovery that was also monitored by the proposed algortihm, and control actions are proposed that lead the BESS to work in normal conditions.