Principal Component Analysis Network Research Articles

Comparison of discriminant analysis coupled with principal component analysis (PCA-DA) and back propagation neural network (BPNN) as classification model for decision support system in porous ceramic matrix (PCM) based burner

Two classification models discriminant analysis coupled with principal component analysis (PCA-DA) and back propagation neural network (BPNN), are for the first time applied for decision support system in porous ceramic matrix (PCM) based burner. The PCM based burner is simulated numerically, and 121 pairs of gas and solid temperature profiles are generated as input data. The operation of PCM based burner is classified into four, on the basis of important properties of PCM like extinction coefficient and convective coupling. With the help of the data, the classification models are developed. The classification models are monitored and analyzed through different plots and classification parameters like specificity, sensitivity and precision. Further, new samples are correctly allocated to their corresponding class by the classification models. The classification models are also explored and compared under noisy data (2% and 5%). The performances of both the classification models are found to be good for no noise case with all the parameters like sensitivity, specificity, and precision values greater than 0.69, for both the models. However, with 2% noise case, BPNN performs better than PCA-DA. The minimum value of parameters (sn, sp, & pr) is 0.67 with BPNN and 0.50 with PCA-DA, respectively. Under 5% noise, the minimum values of the parameters dropped to 0.47 for PCA-DA and 0.50 for BPNN, respectively. With the help of plots though, the new samples are easily identified to their correct class 3.

Robust two-dimensional PCANet with F-norm distance metric

The principal component analysis network (PCANet) is a simplified convolutional neural network (CNN) implemented using a deep subspace learning model. Since its typical PCA filters utilize the squared Frobenius-norm (F-norm) as the distance metric, it exhibits extreme sensitivity to outliers hidden within the data. Furthermore, PCANet faces challenges in extracting structural information in the row and column directions of images as the filters need to be pre-vectorized. To address these issues, we propose a robust bi-directional two-dimensional PCANet (RBDPCANet), wherein the filters utilize the F-norm instead of the squared F-norm to mitigate the impact of outliers. Additionally, the relationship between reconstruction errors and projection distances is simultaneously considered in the objective function. After a concise analytical description of the PCANet concept and related work, we present the methodology for generating three convolutional kernels in the form of RBDPCANet-1, RBDPCANet-2, and RBDPCANet-3 algorithms. These convolution kernels are iteratively obtained and are suitable for numerical implementation. Each of the three algorithms uses another optimization objective function. It has been shown that in image recognition tasks, the developed methods are characterized by better resistance to noise compared to the classic PCANet approach. The rotation invariance and convergence of RBDPCANet are demonstrated according to theoretical principles. The efficacy of the proposed method is validated using images from publicly available datasets. Extensive experimental results on these datasets demonstrate that the proposed RBDPCANet not only significantly boosts the efficiency of image classification but also has strong resistance to occlusions, changes in lighting, changes in orientation, etc. Our implementation code is available at https://github.com/Wesley-li1/RBDPCANet-.

A hybrid EEG classification model using layered cascade deep learning architecture.

The problem of multi-class classification is always a challenge in the field of EEG (electroencephalogram)-based seizure detection. The traditional studies focus on computing or learning a set of features from EEG to distinguish between different patterns. However, the extraction of characteristic information becomes increasingly difficult as the number of EEG types increases. To address this issue, a creative EEG classification technique is proposed by employing a principal component analysis network (PCANet) coupled with phase space reconstruction (PSR) and power spectrum density (PSD). We have introduced the PSR and PSD to prepare the inputs, where dynamic and frequency information are exposed from deep within PCANet. It is remarkable that a layered cascade strategy is designed to make a powerful deep learner according to the rule of one network vs one task (OVO). The proposed method has achieved greater effects than the individual models and shown superior performance in comparison with state-of-the-art algorithms, which present 98.0% of sensitivity, 99.90% of specificity, and 99.07% of accuracy. Our ensemble PCANet model works in an assembly line-like manner, obviating the need for hand-craft features. Results demonstrate that the proposed scheme can greatly enhances the accuracy and robustness of seizure detection from EEG signals.

Review of Predicting Stock Prices Based on Machine Learning and Stock Learning

Predicting stocks is to help investors make decisions and help them minimize risks as much as possible, thereby obtaining investment returns. If we can accurately predict the rise and fall of stock market prices, we can buy or sell stocks at higher prices in the market, thereby obtaining higher profits. This article is based on machine learning and deep learning algorithms such as support vector machine (SVM), principal component analysis (PCA), random forest (RF), and recurrent neural network (RNN), long short-term memory network (LSTM), and convolution neural network (CNN). The conclusion is that machine learning requires manual intervention, is faster to train, more interpretable, and usually performs better on data sets with simpler structures and smaller sizes. Deep learning has lower requirements for feature extraction and discovery, has stronger expression and fitting capabilities, and can better capture the correlation between data in larger and more complex data sets.

Quality Evaluation of Asparagus officinalis by Profile of Amino Acids and Mineral Elements in Different Parts Combined with Chemometrics Methods.

Asparagus officinalis has a homologous value in medicine and vegetables. Its immature stem, commonly called asparagus, is a central edible part. Asparagus skin and leaf also contain rich nutrients. However, these parts are often discarded. This study investigated amino acid and mineral elements in immature stem, skinless asparagus, asparagus skin, and leaf. Their quality was further evaluated by chemometrics methods such as principal component analysis and neural network analysis. The results showed amino acid content was high in immature stem and skinless asparagus and low in leaf, whereas the mineral elements were in four parts. Quality evaluation results showed four parts were divided into three grades. Immature stem and skinless asparagus were grouped into cluster 1 with the best quality as high-quality raw materials in food and health-care products. Meanwhile, three AA (Cys, His, Arg) and two mineral elements (Na, Cr) were identified as quality evaluation iconic substances.

Tracking element-mineral associations with unsupervised learning and dimensionality reduction in chemical and optical image stacks of thin sections

In-situ chemical analysis on thin sections is a cornerstone of geochemical research. Despite massive advances in digital image analysis, the interpretation of such data within the optical petrological context of the thin section has largely remained an analogue task in geochemistry. In this contribution, we registered optical and micro-chemical images from large thin section areas. Chemical datasets from scanning electron microscopy energy dispersive spectrometry (SEM-EDX) and Synchrotron X-ray fluorescence microscopy (S-XFM) were exported from proprietary software. We then evaluated two dimensionality reduction techniques against false-colour images and phase maps to test whether previously unnoticed features became discoverable. Principal component analysis (PCA) and deep sparse autoencoder (DSA) neural networks were used to summarise the multi-channel SEM-EDX and S-XFM into one single red-green-blue (RGB) representation each.Applied to an oceanic gabbro, a cratonic peridotite, and a pelagic limestone, the PCA-based dimensionality reduction was found to produce crisp RGB maps of phases with less noise than false-colour three-element RGB maps. They were registered with optical images to form a simple multi-channel input for pixel-based classification into classes (i.e., semantic segmentation). This was fast and worked well for typical igneous and metamorphic rocks (1–3 h routine). In the chemically quite homogenous limestone, the combined optical and S-XFM PCA input was successfully classified into many phases that were unclassifiable on the conventional SEM-EDX phase map. From the segmented optical and S-XFM PCA map, we exported pixel locations of classified accessory phases back into the original proprietary S-XFM software. This allowed quantification of trace elements whose X-ray peaks were invisible in the original S-XFM quantification.The DSA-based approach yielded phase maps with greater noise, but the technique was very strong at detecting subtle features. In the gabbro sample, the DSA image identified cryptic relict cores in clinopyroxene more clearly than in the previously used Cr concentration map. A DSA training set can be obtained on a small ROI to generate a consistent large area DSA image from which more fragmental cores were observed. For a ROI on the gabbro thin section, the DSA representation map and the blended optical image were segmented and inspected with a loupe tool. This allowed us to generate interactive histograms and bin-scatter concentration plots of selected phases and to search for co-variations of element concentrations. This approach can be expanded to include data from any in situ geochemical acquisition tools and may help discover previously unseen features in complex elemental imagery.

Paternal genetic structure analysis of the modern Han populations based on Y-SNP and Y-STR.

The Han populations represent the largest ethnic group in China. Previous studies have primarily focused on investigating their genetic origins, migration and integration, as well as paternal genetic relationships within specific regional Han populations. However, a comprehensive analysis of the global paternal genetic structure of Han populations is lacking. In this study, we performed Y-chromosome sequencing on 362 unrelated male samples from Chinese Han individuals collected from Qinghai, Sichuan and Liaoning provinces. We then integrated relevant data from reported studies. Our final dataset comprised 1830 samples from 16 Han populations across 15 provinces in China, encompassing information on 89 Y-SNPs and 16 Y-STRs. Statistical analyses were conducted to assess Y-STR haplotype diversity (HD) and Y-SNP haplogroup frequencies. Additionally, we employed principal component analysis (PCA), phylogenetic tree and haplotype network to explore genetic differentiation within Han populations and the genetic relationships between Han populations and ethnic minorities surrounding them. Our results demonstrated that the O-M175 haplogroup represents the predominant paternal lineage in Han populations, with frequencies ranging from 60.53% (Qinghai Han) to 92.7% (Guangdong Han). Moreover, the subclades downstream of O-M175 showed distinct regional variations in their distribution patterns. The O2-M122 haplogroup was prevalent in all Han populations and demonstrated a gradual decline in frequency from north to south. Conversely, the distribution frequency of the O1b-M268 haplogroup decreased from south to north, particularly showed significant presence among Han populations in the Lingnan region. Haplogroup O1a-M119 distributed more frequently in the central Han populations. Our findings revealed that Chinese Han populations can be categorized into three subgroups: northern, central, and southern. Notably, there were significant differences among Han in Qinghai and other regions. Regarding the genetic relationships between Han populations and surrounding ethnic minorities, we observed a closer genetic affinity between different Han populations, but northern Han demonstrated a stronger relationship with the Hui ethnic group, while southern Han exhibited a closer connection with the Gelao and Li ethnic groups. In summary, this study presented a systematic analysis of haplogroup distribution, genetic substructure of Han populations and genetic relationships between Han populations and surrounding ethnic minorities based on 89 Y-SNPs and 16 Y-STRs systematically. Our research supplemented valuable insights into population genetics and forensic genetics, and provided data support for the forensic application of Y chromosome. The integration of Y-SNP haplogroups with Y-STR haplotypes offers enhanced understanding of the genetic substructure within Han populations, which holds significance for both population genetics research and forensic science applications.

A hybrid prediction model of improved bidirectional long short-term memory network for cooling load based on PCANet and attention mechanism

Accurate and reliable cooling load forecasting is a prerequisite for air-conditioning system control and the basis for building-side energy management. Therefore, a hybrid prediction model of an improved bidirectional long short-term memory (BiLSTM) network based on principal component analysis network (PCANet) and attention mechanism (CNN-IBiLSTM-Attention) is proposed to predict the cooling load of large commercial buildings. First of all, the PCANet algorithm is used to analyze the sensitivity of the influencing factors. Then, the hybrid strategy improved whale optimization algorithm (HSIWOA) is used to optimize the hyperparameter of BiLSTM. At last, the performance of the proposed algorithm is verified by using the actual data of two commercial buildings in Xi'an. The results show that using the PCANet algorithm for sensitivity analysis avoids feature redundancy. HSIWOA is suitable for hyperparameter optimization of BiLSTM. Compared with the other three prediction models, CNN-IBiLSTM-Attention reduced the mean absolute percentage error (MAPE) of Building 1 and 2 test sets by 31.55 %, 55.59 %, and 60.58 % and 56.49 %, 60.3 %, and 67.37 %, respectively. The proposed prediction model has superior hyperparameter optimization ability, better model complexity, and stronger generalization ability. Therefore, the proposed prediction model becomes a reliable tool for predicting the cooling load of large commercial buildings.

Artificial neural network-based temperature prediction of a lunar orbiter in thermal vacuum test: Data-driven reduced-order models

This study presents data-driven reduced-order models (ROMs) of a lunar orbiter based on principal component analysis (PCA) and artificial neural networks (ANNs) for a ground thermal vacuum test to simulate space thermal environments. We employed a radial basis function network (RBFN) and deep neural network (DNN) from among the various types of ANNs. PCA extracts features from high-dimensional data, such as thermal analysis data. It is utilized in machine learning algorithms as a preprocessing step before inputting the data into neural networks. This process improves the convergence speed and training performances compared to using neural networks alone. The coefficients of the extracted principal component modes were regressed using the RBFN and DNN. Twenty thermal design parameters comprising infrared emissivity, effective thermal conductivity, thermal contact conductance coefficients, and thermal conductance were used to train the ROMs. We conducted training and test of the proposed models during the cold and hot balance phases of the ground test. Consequently, the temperature map can be estimated in seconds for the new design parameters, and the model results are consistent with thermal analysis and measurement data.

Seismic fault identification in coal mines based on the self-organizing map-gray wolf optimizer-support vector machine algorithm

Accurate fault identification in coal mines is important to improve mine safety and economic benefits. We compare various intelligent algorithms for data preprocessing and optimization and analyze the construction methods of seismic attribute data sets and the performance of intelligent optimization algorithms using fault identification accuracy as the discrimination index to find a better combined model for seismic fault identification. First, the training data set is constructed by mining the fault and nonfault information revealed by the roadway. The distribution characteristics of the seismic attribute data indicate similarities among them, and they are nonlinearly separable. Directly using the attributes to construct the data set, the accuracy of fault identification using the support vector machine (SVM) model is 78.41%. Principal component analysis (PCA) and self-organizing mapping (SOM) neural networks are used to extract effective information and then combined with the SVM classification model, and the accuracy of fault identification is 83.82% and 87.47%, respectively. Compared with the original data and PCA dimensionality reduction data, the accuracy of fault detection is improved by 9.06% and 3.66%, respectively, indicating that SOM can effectively improve the accuracy of fault detection by eliminating similar attributes and reducing the weight of redundant information. Then, through a fixed attribute data set, genetic algorithm (GA), particle swarm optimization (PSO), and gray wolf optimizer (GWO) intelligent optimization algorithms are used to find the optimal kernel function parameter and penalty parameter of the SVM classifier. The accuracy rate of the SOM-GWO-SVM model reaches 91.12%, compared with the SOM-PSO-SVM and SOM-GA-SVM, and the model accuracy is increased by 5.2% and 5.61%, respectively. Compared with PSO and GA, the GWO algorithm has a better global search ability. The identification result of the SOM-GWO-SVM model is closest to the actual fault exposure, especially for the identification of “short” faults and associated faults, which has obvious advantages over the traditional manual interpretation in terms of efficiency and accuracy.

Functional PCA and deep neural networks-based Bayesian inverse uncertainty quantification with transient experimental data

This work focuses on developing an inverse uncertainty quantification (IUQ) process for time-dependent responses, using dimensionality reduction by functional principal component analysis (PCA) and deep neural network (DNN)-based surrogate models. The demonstration is based on the IUQ of TRACE physical model parameters using the FEBA benchmark transient experimental data on peak cladding temperatures during reflooding. The quantity-of-interest (QoI) is time-dependent peak cladding temperature (PCT) profiles. Conventional PCA can hardly represent the data precisely due to the sudden temperature drop at the time of quenching. As a result, a functional alignment method is used to separate the phase and amplitude information in the PCT profiles before conventional PCA is applied for dimensionality reduction. The resulting PC scores are then used to build DNN-based surrogate models to significantly reduce the computational cost in Markov Chain Monte Carlo sampling, while the code/interpolation uncertainty is accounted for using Bayesian neural networks. We compared four IUQ processes with different dimensionality reduction methods and surrogate models. The proposed approach has demonstrated the best performance in reducing the dimensionality of the transient PCT profiles. This approach also produces the posterior distributions of the physical model parameters with which the model predictions have the best agreement with the experimental data.

Joint principal component analysis and BP neural network to measure the development level of modern logistics in China

Joint principal component analysis and BP neural network to measure the development level of modern logistics in China

Fusion PCAM R-CNN of Automatic Segmentation for Magnetic Flux Leakage Defects.

Magnetic leakage detection technology plays an important role in the long-oil pipeline. Automatic segmentation of defecting images is crucial for the detection of magnetic flux leakage (MFL) works. At present, accurate segmentation for small defects has always been a difficult problem. In contrast to the state-of-the-art MFL detection methodologies based on convolution neural network (CNN), an optimization method is devised in our study by integrating mask region-based CNN (Mask R-CNN) and information entropy constraint (IEC). To be precise, the principal component analysis (PCA) is utilized to improve the feature learning and network segmentation ability of the convolution kernel. The similarity constraint rule of information entropy is proposed to be inserted into the convolution layer in the Mask R-CNN network. The Mask R-CNN optimizes the convolutional kernel with similar weights or higher similarity, meanwhile, the PCA network reduces the dimension of the feature image to reconstruct the original feature vector. As such, the feature extraction of MFL defects is optimized in the convolution check. The research results can be applied in the field of MFL detection.

Evaluation of mine water quality based on the PCA–PSO–BP model

Abstract To enhance the mining area's overall use of mine water in the arid area of Western China and mitigate the current water scarcity problem, this paper introduces an intelligent optimization algorithm and neural network for mine water quality evaluation and proposes a principal component analysis (PCA)–particle swarm optimization (PSO)–back propagation (BP) mine water quality evaluation model. Firstly, the model uses PCA to identify the primary factors affecting mine water quality, then enhances the optimal weights and thresholds of the BP neural network based on the PSO algorithm, and the PCA–PSO–BP evaluation model with nine input layers, nine hidden layers, and one output layer is created. In addition, using the Shicaocun Mine as an example, the results demonstrate that the PCA–PSO–BP model has accurate mine water quality evaluation results, and the prediction accuracy reached 86.8255%. This exemplifies the PSO method's superiority to the BP neural network improvement. This study not only offers a novel theoretical framework for assessing and forecasting water quality in mining regions, but it also sets the stage for the possible broad use of state-of-the-art neural networks and optimization algorithms in the coal mining industry.

Data driven reduced modeling for fluidized bed with immersed tubes based on PCA and Bi-LSTM neural networks

The fast and accurate reduced-order modeling of fluidized beds is a challenging task in the field of fluid dynamics, owing to their high dimensionality and nonlinear dynamic behavior. In this study, a nonintrusive reduced order modeling method, the reduced order model based on principal component analysis and bidirectional long short-term memory networks (PBLSTM ROM), was developed to capture complex spatio-temporal dynamics of fluidized beds. By combining principal component analysis and Bidirectional long- short-term memory networks, the PBLSTM ROM effectively extracted dynamic evolution information without any prior knowledge of governing equations, enabling reduced-order modeling of unsteady flow fields. The PBLSTM ROM was validated using the solid volume fraction and gas velocity flow fields of a fluidized bed with immersed tubes, showing superior performance over both the PLSTM and PANN ROMs in accurately capturing temporal changes in the fluidization fields, especially in the region near immersed tubes where severe fluctuations appear. Moreover, the PBLSTM ROM improved the simulation speed by five orders of magnitude compared to traditional computational fluid dynamics simulations. These findings suggest that the PBLSTM ROM presents a promising approach for analyzing the complex fluid flows in engineering practice.

Cancelable face recognition using phase retrieval and complex principal component analysis network

Cancelable face recognition using phase retrieval and complex principal component analysis network

HPLC-LED-Induced Fluorescence Analysis of Tear Fluids: An Objective Method for Primary Open Angle Glaucoma Diagnosis

PurposeThe study showcased the application of the lab-assembled HPLC-LED-IF system to analyze proteins in tear fluid samples collected from individuals diagnosed with primary open-angle glaucoma (POAG).MethodsClinical application of the said technique was evaluated by recording chromatograms of tear fluid samples from control and POAG subjects and by analyzing the protein profile using multivariate analysis. The data analysis methods involved are principal component analysis (PCA), Match/No-Match, and artificial neural network (ANN) based binary classification for disease diagnosis.ResultsMahalanobis distance and spectral residual values calculated using a standard calibration set of clinically confirmed POAG samples for the Match/No-Match test gave 86.9% sensitivity and 81.8% specificity. ANN with leaving one out procedure has given 87.1% sensitivity and 81.8% specificity.ConclusionThe results of the study revealed that the utilization of a 278 nm LED excitation in the HPLC system offers good sensitivity for detecting proteins at low concentrations allowing to obtain reliable protein profiles for the diagnosis of POAG.

Development of machine-learning-based models for identifying the sources of nitrate and fluoride in groundwater and predicting their human health risks

AbstractThis research aimed to identify the main sources of groundwater pollution and assess the non-carcinogenic human health risk resulting from nitrate and fluoride contamination. These goals were achieved by employing unsupervised and supervised machine algorithms, including principal component analysis (PCA) and multilayer perceptron artificial neural networks (MLP-ANN). Thirty-seven groundwater samples were analyzed for twelve physical and chemical parameters, including pH, EC, TDS, TH, Cl, F, SO4, NO3, Ca, Mg, Na, and HCO3, and the initial investigation indicated that except for Cl, F, Ca, and Mg, all the parameters are above the guidelines of the World Health Organization (WHO). PCA indicated that mineral dissolution is the main source of F, while high NO3 concentration primarily resulted from agricultural operation due to extensive use of nitrogen and calcium-based fertilizers. Consequently, the non-carcinogenic human health risk (HHR) for children and adults is evaluated based on NO3 and F. The conventional approach for assessing HHR is time-consuming and often associated with errors in calculating hazard quotients (HQ) and hazard indices (HI). In this research, MLP-ANN is suggested to overcome these limitations. In the MLP-ANN modeling, the data were divided into two parts training (80%) and validation (20%), with NO3 and F concentration as inputs and HQ and HI as outputs. The performance of the resulting models was tested using root mean square error (RMSE) and coefficient of determination (R2). The model provided a satisfactory result with a maximum RMSE of 4% and R2 higher than 97% for training and validation. As a result, obtained HIs suggested that 97.3% of the groundwater samples in the study area are suitable for human consumption. The non-carcinogenic HHR is successfully assessed using machine learning algorithms, and the results have led to the conclusion that this approach is highly recommended for effectively managing groundwater resources.

Impact of exposure temperature rise on mass mortality of tidal flat pacific oysters

Oysters are a major commercial and ecological fishery resource. Recently, the oyster industry has experienced mass mortality in summer due to environmental factors. Generally, the survival of oysters in aquatic environments is mainly impacted by environmental stressors such as elevated sea temperatures and reduced salinity; however, the stressors impacting tidal flat oysters that are repeatedly exposed to air remain poorly understood. Hence, we studied the relationship between environmental factors and the survival of tidal flat pacific oysters in Incheon, South Korea, where mass mortality is common. Principal component analysis and Bayesian networks revealed that air temperature (in spring and summer) and sea temperature (in summer) are related to oyster production. In habitats of the tidal flat oysters during the summer, high temperatures of 34.7–35.4°C (maximum 47.6°C) were observed for average durations of 0.8–1.9 hours (maximum 3.6 hours). Furthermore, heat waves occurred for up to 12 consecutive days. Results from the multiple stress test showed that when exposed to 45°C (air temperature) for 4 hours per day, the survival rate of oysters was 42.5% after only 2 days and 0% after 6 days. The findings stemming from the field observations and stress tests suggest that high temperatures during emersion may contribute to mass mortality of oysters in summer, indicating a potential threat to oysters due to climate change. To understand the effects of future thermal stress on oysters more accurately, simultaneous long-term trend analyses and field-based observations are required.

Monitoring of polycyclic aromatic hydrocarbons and toxic metals in blue mussels and oysters for human consumption from Danish waters (2005–2019)

Bivalve molluscs have for many years been used as indicator organisms for environmental pollution with various chemical substances. However, limited information is available on the levels of chemical contaminants in mussels and oysters intended for human consumption. Polycyclic aromatic hydrocarbons (PAHs) and toxic metals, such as cadmium (Cd), mercury (Hg) and lead (Pb) have been analysed in edible bivalve samples and the present study provides the results from samples collected in Danish waters from 2005 to 2019. Principal Component Analysis (PCA) and Pattern Recognition Neural Network (PRNN) analysis were applied to data to identify any differences according to organism, locality, and contaminant levels. Differences were confirmed for Pb concentrations being generally higher in mussels than oysters. Based on toxic metals results, the PCA and PRNN analyses made it possible to categorise samples as a mussel or an oyster, as well as make decisions on the origin of a specific sample. All concentrations in mussels and oysters were below the current European Union maximum levels for the contaminants.