Relevant Principal Components Research Articles

Patterns and correlations in European electricity prices.

Electricity markets are central to the coordination of power generation and demand. The European power system is divided into several bidding zones, each having an individual electricity market price. While individual price time series have been intensively studied in recent years, spatiotemporal aspects have received little attention. This article provides a comprehensive data-centric analysis of the patterns and correlations of the European day-ahead electricity prices between 2019 and 2023, characteristically abnormal due to the energy crisis in Europe. We identify the dominant communities of bidding zones and show that spatial differences can be described with very few principal components. Most bidding zones in Continental Europe were brought together during the energy crisis: Correlations increased, and the number of relevant principal components decreased. Opposite effects occur in the Nordic countries and the Iberian Peninsula where correlations decrease and communities fragment.

Principal Component Analysis for Distributions Observed by Samples in Bayes Spaces

Distributional data have recently become increasingly important for understanding processes in the geosciences, thanks to the establishment of cost-efficient analytical instruments capable of measuring properties over large numbers of particles, grains or crystals in a sample. Functional data analysis allows the direct application of multivariate methods, such as principal component analysis, to such distributions. However, these are often observed in the form of samples, and thus incur a sampling error. This additional sampling error changes the properties of the multivariate variance and thus the number of relevant principal components and their direction. The result of the principal component analysis becomes an artifact of the sampling error and can negatively affect the subsequent data analysis. This work presents a way of estimating this sampling error and how to confront it in the context of principal component analysis, where the principal components are obtained as a linear combination of elements of a newly constructed orthogonal spline basis. The effect of the sampling error and the effectiveness of the correction is demonstrated with a series of simulations. It is shown how the interpretability and reproducibility of the principal components improve and become independent of the selection of the basis. The proposed method is then applied on a dataset of grain size distributions in a geometallurgical dataset from Thaba mine in the Bushveld complex.

The impact of the concept of “tailor-made teaching” on the cultivation of higher vocational art and design professionals in the era of artificial intelligence

Abstract This paper first studied the education concept of tailor-made teaching, explored the influence of tailor-made teaching on the cultivation of professional ability, and constructed the talent cultivation model of art majors based on tailor-made teaching. Then, we constructed the influencing factor analysis model, extracted the relevant principal components, calculated the Spearman correlation coefficient, established the multiple linear regression analysis model, estimated the model parameters, and tested the model. Finally, we analyzed the influence of the concept of tailor-made teaching on talent cultivation mode, cultivation effect and cultivation evaluation, and carried out correlation analysis, regression analysis and mediation effect test on the influencing factors. The results show that the mean value of the evaluation of the education model of “tailor-made education” ranges from 3.78 to 4.35, which is at a relatively satisfactory level. The sig value of the relationship between “teaching according to student’s abilities” and the quality of talent training is 0.000, indicating that “teaching according to student’s abilities” has a significant positive effect on the quality of talent training.

VGG16 feature selection using PCA-big bang big algorithm

In the recent decade, plant disease classification using convolution neural networks has proven to be superior because of its ability to extract key features. Obtaining the optimum feature subset with the necessary discriminant information is challenging. The main objective of this paper is to design an efficient hybrid plant disease feature selection approach and validate it on standard image datasets. The raw input image features were transformed into 8192 learned features by employing the VGG16. To reduce the training time and enhance classification accuracy, the dimensionality reduction technique Principal Component Analysis (PCA) is integrated with the big bang-big crunch (BBBC) optimization algorithm. The PCA-BBBC feature selection method reduces computing time by eliminating unnecessary and redundant features. The proposed approach was evaluated on plant diseases and benchmarked image datasets. Experimental results reveal that the Artificial Neural Network (ANN) classifier integrated with the VGG16-PCA-BBBC approach enhanced the performance of the classifier. The proposed approach outperformed the VGG16-PCA-ANN method and other popular image classification techniques. For the rice disease dataset, the proposed hybrid approach reduced the VGG16 extracted 8192 deep features to 200 relevant principal components. The recommended reduced features were used for training ANN. The test dataset was classified by ANN with an accuracy of 99.12%. Experimental results demonstrate that the proposed approach improved the performance of the classifier and accurately labeled image and plant diseases datasets aiding farmers to adopt remedial measures.

Automatic monitoring and detection of tail-biting behavior in groups of pigs using video-based deep learning methods.

Automated monitoring of pigs for timely detection of changes in behavior and the onset of tail biting might enable farmers to take immediate management actions, and thus decrease health and welfare issues on-farm. Our goal was to develop computer vision-based methods to detect tail biting in pigs using a convolutional neural network (CNN) to extract spatial information, combined with secondary networks accounting for temporal information. Two secondary frameworks were utilized, being a long short-term memory (LSTM) network applied to sequences of image features (CNN-LSTM), and a CNN applied to image representations of sequences (CNN-CNN). To achieve our goal, this study aimed to answer the following questions: (a) Can the methods detect tail biting from video recordings of entire pens? (b) Can we utilize principal component analyses (PCA) to reduce the dimensionality of the feature vector and only use relevant principal components (PC)? (c) Is there potential to increase performance in optimizing the threshold for class separation of the predicted probabilities of the outcome? (d) What is the performance of the methods with respect to each other? The study utilized one-hour video recordings of 10 pens with pigs prior to weaning, containing a total of 208 tail-biting events of varying lengths. The pre-trained VGG-16 was used to extract spatial features from the data, which were subsequently pre-processed and divided into train/test sets before input to the LSTM/CNN. The performance of the methods regarding data pre-processing and model building was systematically compared using cross-validation. Final models were run with optimal settings and evaluated on an independent test-set. The proposed methods detected tail biting with a major-mean accuracy (MMA) of 71.3 and 64.7% for the CNN-LSTM and the CNN-CNN network, respectively. Applying PCA and using a limited number of PCs significantly increased the performance of both methods, while optimizing the threshold for class separation did result in a consistent but not significant increase of the performance. Both methods can detect tail biting from video data, but the CNN-LSTM was superior in generalizing when evaluated on new data, i.e., data not used for training the models, compared to the CNN-CNN method.

Fully component selection: An efficient combination of feature selection and principal component analysis to increase model performance

Principal component analysis (PCA) is one the most effective and widely used dimensionality-reduction techniques which aggregates the maximal variance in the first few components. In many applications, the first few components of PCA are used in place of the original variables in statistical and machine learning models; nevertheless, such use does not guarantee the selection of the most relevant components to the target variable. This paper presents an efficient approach in which all components, rather than simply the first few, are considered as input for the random forest (RF) model, and a feature selection algorithm is integrated with the RF to select the most relevant components, hence called fully component selection (FCS). The proposed method was evaluated on spectroscopic data comprising 70 soil samples with 2050 spectral bands to estimate soil organic carbon (SOC). The FCS approach was compared to the RF model once considering the original features, then again using only the first few components, and again using all components without feature selection. The results showed that RF-FCS substantially outperformed other approaches, such that the R2 was increased between 25.2% and 55.5%. Furthermore, the findings indicated that the most relevant principal components (PCs) to the target variable were not the first few, but PC6, PC7, PC8, PC15, and PC43. Using the FCS approach increased model performance substantively, and coupling it with machine learning and statistical models is strongly recommended for high dimensional data applications.

Adaptive denoising for chemical exchange saturation transfer MR imaging.

High image signal-to-noise ratio (SNR) is required to reliably detect the inherently small chemical exchange saturation transfer (CEST) effects in vivo. In this study, it was demonstrated that identifying spectral redundancies of CEST data by principal component analysis (PCA) in combination with an appropriate data-driven extraction of relevant information can be used for an effective and robust denoising of CEST spectra. The relationship between the number of relevant principal components and SNR was studied on fitted in vivo Z-spectra with artificially introduced noise. Three different data-driven criteria to automatically determine the optimal number of necessary components were investigated. In addition, these criteria facilitate straightforward assessment of data quality that could provide guidance for CEST MR protocols in terms of SNR. Insights were applied to achieve a robust denoising of highly sampled low power Z-spectra of the human brain at 3 and 7T. The median criterion provided the best estimation for the optimal number of components consistently for all three investigated artificial noise levels. Application of the denoising technique to in vivo data revealed a considerable increase in image quality for the amide and rNOE contrast with a considerable SNR gain. At 7T the denoising capability was quantified to be comparable or even superior to an averaging of six measurements. The proposed denoising algorithm enables an efficient and robust denoising of CEST data by combining PCA with appropriate data-driven truncation criteria. With this generally applicable technique at hand, small CEST effects can be reliably detected without the need for repeated measurements.

Comparison of multi-response prediction methods

While data science is battling to extract information from the enormous explosion of data, many estimators and algorithms are being developed for better prediction. Researchers and data scientists often introduce new methods and evaluate them based on various aspects of data. However, studies on the impact of/on a model with multiple response variables are limited. This study compares some newly-developed (envelope) and well-established (PLS, PCR) prediction methods based on real data and simulated data specifically designed by varying properties such as multicollinearity, the correlation between multiple responses and position of relevant principal components of predictors. This study aims to give some insight into these methods and help the researcher to understand and use them in further studies.

Relative Principal Components Analysis: Application to Analyzing Biomolecular Conformational Changes

A new method termed “Relative Principal Components Analysis” (RPCA) is introduced that extracts optimal relevant principal components to describe the change between two data samples representing two macroscopic states. The method is widely applicable in data-driven science. Calculating the components is based on a physical framework that introduces the objective function (the Kullback–Leibler divergence) appropriate for quantifying the change of the macroscopic state affected by the changes in the microscopic features. To demonstrate the applicability of RPCA, we analyze the thermodynamically relevant conformational changes of the protein HIV-1 protease upon binding to different drug molecules. In this case, the RPCA method provides a sound thermodynamic foundation for analyzing the binding process and thus characterizing both the collective and the locally relevant conformational changes. Moreover, the relevant collective conformational changes can be reconstructed from the informative latent variables to exhibit both the enhanced and the restricted conformational fluctuations upon ligand association.

Classification of Romanian medicinal plant extracts according to the therapeutic effects using thin layer chromatography and robust chemometrics

Thin layer chromatography in combination with image analysis and advanced chemometric methods were successfully used to classify the medicinal herbs according to their therapeutic effects and usage. The investigations were conducted using two types of plates (HPTLC Silica gel 60 and HPTLC Silica gel 60 F254) which were evaluated in UV light at 254 and 365 nm. The holistic evaluation of the numerical data corresponding different image processing channels (blue, grey, red, green) was performed by employing appropriate multivariate methods: hierarchical cluster analysis (HCA), principal component analysis (PCA), fuzzy principal component analysis (FPCA) and linear discriminant analysis (LDA) applied to the first relevant principal components. The results obtained by applying LDA method indicate a highly accurate separation of the medicinal herbs within the four groups, in good agreement with therapeutic effects and usage. According to this classification, the best image processing channels were identified for each of the investigated HPTLC plates: blue channel for HPTLC Silica gel 60 F254 (with 92.9% percent of discrimination in case of PCA and FPCA) and respectively red channel for HPTLC Silica gel 60 (with 93.9% percent of discrimination in case of FPCA). The 2D and 3D score scatterplots illustrate also the accurate and reliable discrimination between the four distinct groups.

Characterization of Clinically Relevant Fungi via SERS Fingerprinting Assisted by Novel Chemometric Models.

Nonculture-based tests are gaining popularity and upsurge in the diagnosis of invasive fungal infections (IFI) fostered by their main asset, the reduced analysis time, which enables a more rapid diagnosis. In this project, three different clinical isolates of relevant filamentous fungal species were discriminated by using a rapid (less than 5 min) and sensitive surface-enhanced Raman scattering (SERS)-based detection method, assisted by chemometrics. The holistic evaluation of the SERS spectra was performed by employing appropriate chemometric tools-classical and fuzzy principal component analysis (FPCA) in combination with linear discriminant analysis (LDA) applied to the first relevant principal components. The efficiency of the proposed robust algorithm is illustrated on the data set including three fungal isolates (Aspergillus fumigatus sensu stricto, cryptic A. fumigatus complex species, and Rhizomucor pusillus) that were isolated from patient materials. The accurate and reliable discrimination between species of common fungal pathogen strains suggest that the developed method has the potential as an alternative, spectroscopic-based routine analysis tool in IFI diagnosis.

Abstract WP204: Genetic Variant in VCAM1 Mediates Acute Infarct Size in Ischemic Stroke Patients

Introduction: Even though neuroinflammation is increasingly recognized as an essential contributor to ischemic brain injury, the exact underlying mechanisms remain unclear. Higher expression of the vascular cell adhesion molecule 1 (VCAM-1) increases leukocyte-brain endothelium interaction and has been associated with larger infarct size following acute ischemic stroke (AIS). The activation of the TGF-β signaling pathways can down-regulate VCAM-1 expression and ameliorate deleterious tissue outcome during neuroinflammation. Hypothesis: We sought to investigate whether genetic variation in the TGF-beta pathway and adhesion molecule genes is associated with acute stroke lesion size. Methods: We completed genome-wide association (GWA) testing and diffusion-weighted imaging lesion volume (DWIv) analysis in a discovery cohort of 532 AIS patients of European ancestry enrolled within 48 hours of symptom onset. An independent European ancestry cohort of 724 AIS patients with GWA data and automated DWIv served as replication cohort. GWA testing per SNP was performed using linear regression modeling of natural log-transformed DWIv adjusted for age, sex and relevant principal components. We selected 42 inflammatory genes in the TGF-beta pathway for a gene-based analysis using VEGAS (Versatile Gene-based -Association Study) software. A pre-specified discovery phase Bonferroni-corrected threshold was set at p<0.001. In the replication phase, 14 SNPs overlapping were tested at the Bonferroni-corrected p-value threshold of p<0.004. Results: Of all genes in the TGF-beta pathway, VCAM1 (p=0.0006) was significantly associated with DWIv in the discovery AIS cohort (age: 66 ± 14.9 years, sex: 63.4% male, DWIv: 2.2cm 3 (IQR: 0.6-11.7cm 3 )). A single SNP within the VCAM1 gene boundaries (rs3176876 (BETA=0.2341, p=0.003)) was significantly associated with DWIv in the replication AIS cohort (age: 65 ± 14.1 years, sex: 68.6% male, DWIv: 4cm 3 (IQR: 1.3-17.2cm 3 )). Conclusion: The genetic variant rs3176876 in the VCAM1 gene is associated with larger infarct lesion size measured on brain MRI of AIS patients. These findings suggest genetic contribution to the pro-inflammatory mechanisms in acute cerebral ischemia and warrant further investigation.

Unsupervised change detection in VHR remote sensing imagery – an object-based clustering approach in a dynamic urban environment

Monitoring of changes is one of the most important inherent capabilities of remote sensing. The steadily increasing amount of available very-high resolution (VHR) remote sensing imagery requires highly automatic methods and thus, largely unsupervised concepts for change detection. In addition, new procedures that address this challenge should be capable of handling remote sensing data acquired by different sensors. Thereby, especially in rapidly changing complex urban environments, the high level of detail present in VHR data indicates the deployment of object-based concepts for change detection. This paper presents a novel object-based approach for unsupervised change detection with focus on individual buildings. First, a principal component analysis together with a unique procedure for determination of the number of relevant principal components is performed as a predecessor for change detection. Second, k-means clustering is applied for discrimination of changed and unchanged buildings. In this manner, several groups of object-based difference features that can be derived from multi-temporal VHR data are evaluated regarding their discriminative properties for change detection. In addition, the influence of deviating viewing geometries when using VHR data acquired by different sensors is quantified. Overall, the proposed workflow returned viable results in the order of κ statistics of 0.8–0.9 and beyond for different groups of features, which demonstrates its suitability for unsupervised change detection in dynamic urban environments. With respect to imagery from different sensors, deviating viewing geometries were found to deteriorate the change detection result only slightly in the order of up to 0.04 according to κ statistics, which underlines the robustness of the proposed approach.

Integrative clustering of high-dimensional data with joint and individual clusters.

When measuring a range of genomic, epigenomic, and transcriptomic variables for the same tissue sample, an integrative approach to analysis can strengthen inference and lead to new insights. This is also the case when clustering patient samples, and several integrative cluster procedures have been proposed. Common for these methodologies is the restriction to a joint cluster structure, equal in all data layers. We instead present a clustering extension of the Joint and Individual Variance Explained algorithm (JIVE), Joint and Individual Clustering (JIC), enabling the construction of both joint and data type-specific clusters simultaneously. The procedure builds on the connection between k-means clustering and principal component analysis, and hence, the number of clusters can be determined by the number of relevant principal components. The proposed procedure is compared with iCluster, a method restricted to only joint clusters, and simulations show that JIC is clearly advantageous when both individual and joint clusters are present. The procedure is illustrated using gene expression and miRNA levels measured in breast cancer tissue from The Cancer Genome Atlas. The analysis suggests a division into three joint clusters common for both data types and two expression-specific clusters.

Factor models for alpha streams

We propose a framework for constructing factor models for alpha streams. Our motivation is threefold. Firstly, when the number of alphas is large, the sample covariance matrix is singular. Secondly, out-of-sample stability of the sample covariance matrix is challenging. We discuss various risk factors for alphas, such as style risk factors, cluster risk factors based on alpha taxonomy, principal components and using the underlying tradables (stocks) as alpha risk factors, for which computing the factor loadings and factor covariance matrixes does not involve any correlations with alphas, and their number is much larger than that of the relevant principal components. We draw insight from stock factor models, but also point out substantial differences.

Comparison of correlation ranking and eigenvalue ranking unfolded principal component regression for direct determination of naproxen in human serum using excitation–emission matrix fluorescence spectroscopy

Using principal components as input of a multivariate calibration model always brings up the question what subset of PCs gives the best reliable regression model with the most predictability. In this current study, two different methods are presented and discussed for the determination of naproxen (NAP) in serum based on principal component regression. At the first step, the fluorescence landscapes of NAP in serum with excitation wavelengths from 235 to 293 nm and emission wavelengths in the range 320–420 nm were obtained. The resulted excitation emission fluorescence matrices were unfolded and subjected to principal component analysis, and a linear regression model was built to model the relationship between the extracted principal components (PCs) and the concentrations. Two approaches were applied to select the most relevant principal components: eigenvalue ranking and correlation ranking. The results of this study showed that the importance of each PC to the multivariate calibration is not correctly described by its eigenvalue, but by its influence on the prediction of the dependent variable. Since it is possible for a PC to have a low variance in original dataset (low eigenvalue) but a high correlation with the dependent variable, eigenvalue ranking selection should not be employed. The proposed method is an interesting alternative to the traditional techniques normally used for determining naproxen such as HPLC. Compared to HPLC method, the proposed EV-PCR and CR-PCR methods were rapid, easy and of low cost for the quantification of NAP in serum using simple UV–Vis spectrofluorimetry.

Clinical study of noninvasive in vivo melanoma and nonmelanoma skin cancers using multimodal spectral diagnosis.

The goal of this study was to determine the diagnostic capability of a multimodal spectral diagnosis (SD) for in vivo noninvasive disease diagnosis of melanoma and nonmelanoma skin cancers. We acquired reflectance, fluorescence, and Raman spectra from 137 lesions in 76 patients using custom-built optical fiber-based clinical systems. Biopsies of lesions were classified using standard histopathology as malignant melanoma (MM), nonmelanoma pigmented lesion (PL), basal cell carcinoma (BCC), actinic keratosis (AK), and squamous cell carcinoma (SCC). Spectral data were analyzed using principal component analysis. Using multiple diagnostically relevant principal components, we built leave-one-out logistic regression classifiers. Classification results were compared with histopathology of the lesion. Sensitivity/specificity for classifying MM versus PL (12 versus 17 lesions) was 100%/100%, for SCC and BCC versus AK (57 versus 14 lesions) was 95%/71%, and for AK and SCC and BCC versus normal skin (71 versus 71 lesions) was 90%/85%. The best classification for nonmelanoma skin cancers required multiple modalities; however, the best melanoma classification occurred with Raman spectroscopy alone. The high diagnostic accuracy for classifying both melanoma and nonmelanoma skin cancer lesions demonstrates the potential for SD as a clinical diagnostic device.

Factor Models for Alpha Streams

We propose a framework for constructing factor models for alpha streams. Our motivation is threefold. 1) When the number of alphas is large, the sample covariance matrix is singular. 2) Its out-of-sample stability is challenging. 3) Optimization of investment allocation into alpha streams can be tractable for a factor model alpha covariance matrix. We discuss various risk factors for alphas such as: style risk factors; cluster risk factors based on alpha taxonomy; principal components; and also using the underlying tradables (stocks) as alpha risk factors, for which computing the factor loadings and factor covariance matrices does not involve any correlations with alphas, and their number is much larger than that of the relevant principal components. We draw insight from stock factor models, but also point out substantial differences.

Towards a More Nuanced View of Vocal Attractiveness

This study reports on male and female Californians' ratings of vocal attractiveness for 30 male and 30 female voices reading isolated words. While ratings by both sexes were highly correlated, males generally rated fellow males as less attractive than females did, but both females and males had similar ratings of female voices. Detailed acoustic analyses of multiple parameters followed by principal component analyses on vowel and voice quality measures were conducted. Relevant principal components, along with additional independent acoustic measures, were entered into regression models to assess which acoustic properties predict attractiveness ratings. These models suggest that a constellation of acoustic features which indicate apparent talker size and conformity to community speech norms contribute to perceived vocal attractiveness. These results suggest that judgments of vocal attractiveness are more complex than previously described.

Multimodal MR imaging model to predict tumor infiltration in patients with gliomas

Gliomas remain difficult to treat, in part, due to our inability to accurately delineate the margins of the tumor. The goal of our study was to evaluate if a combination of advanced MR imaging techniques and a multimodal imaging model could be used to predict tumor infiltration in patients with diffuse gliomas. Institutional review board approval and written consent were obtained. This prospective pilot study enrolled patients undergoing stereotactic biopsy for a suspected de novo glioma. Stereotactic biopsy coordinates were coregistered with multiple standard and advanced neuroimaging sequences in 10 patients. Objective imaging values were assigned to the biopsy sites for each of the imaging sequences. A principal component analysis was performed to reduce the dimensionality of the imaging dataset without losing important information. A univariate analysis was performed to identify the statistically relevant principal components. Finally, a multivariate analysis was used to build the final model describing nuclear density. A univariate analysis identified three principal components as being linearly associated with the observed nuclear density (p values 0.021, 0.016, and 0.046, respectively). These three principal component composite scores are predominantly comprised of DTI (mean diffusivity or average diffusion coefficient and fractional anisotropy) and PWI data (rMTT, Ktrans). The p value of the model was <0.001. The correlation between the predicted and observed nuclear density was 0.75. A multi-input, single output imaging model may predict the extent of glioma invasion with significant correlation with histopathology.