Statistical Pattern Recognition Techniques Research Articles

Connecting Obsidian Artifacts with Their Sources Using Multivariate Statistical Analysis of LIBS Spectral Signatures

With the recent introduction of handheld instruments for field use, laser-induced breakdown spectroscopy (LIBS) is emerging as a practical technology for real-time in situ geochemical analysis in the field. LIBS is a form of optical emission spectroscopy that is simultaneously sensitive to all elements with a single laser shot so that a broadband LIBS spectrum can be considered a diagnostic geochemical fingerprint. Sets of LIBS spectra were collected for seven obsidian centers across north-central California, with data processed using multivariate statistical analysis and pattern recognition techniques. Although all obsidians exhibit similar bulk compositions, different regional obsidian sources were effectively discriminated via partial least squares discriminant analysis. Obsidian artifacts from seven archaeological sites were matched to their putative sources with a high degree of confidence.

Advances and challenges in the detection of academic stress and anxiety in the classroom: A literature review and recommendations.

In recent years, stress and anxiety have been identified as two of the leading causes of academic underachievement and dropout. However, there is little work on the detection of stress and anxiety in academic settings and/or its impact on the performance of undergraduate students. Moreover, there is a gap in the literature in terms of identifying any computing, information technologies, or technological platforms that help educational institutions to identify students with mental health problems. This paper aims to systematically review the literature to identify the advances, limitations, challenges, and possible lines of research for detecting academic stress and anxiety in the classroom. Forty-four recent articles on the topic of detecting stress and anxiety in academic settings were analyzed. The results show that the main tools used for detecting anxiety and stress are psychological instruments such as self-questionnaires. The second most used method is acquiring and analyzing biological signals and biomarkers using commercial measurement instruments. Data analysis is mainly performed using descriptive statistical tools and pattern recognition techniques. Specifically, physiological signals are combined with classification algorithms. The results of this method for detecting anxiety and academic stress in students are encouraging. Using physiological signals reduces some of the limitations of psychological instruments, such as response time and self-report bias. Finally, the main challenge in the detection of academic anxiety and stress is to bring detection systems into the classroom. Doing so, requires the use of non-invasive sensors and wearable systems to reduce the intrinsic stress caused by instrumentation.

Applications of Electronic Nose Coupled with Statistical and Intelligent Pattern Recognition Techniques for Monitoring Tea Quality: A Review

Tea is the most widely consumed non-alcoholic beverage worldwide. In the tea sector, the high demand for tea has led to an increase in the adulteration of superior tea grades. The procedure of evaluating tea quality is difficult to assure the highest degree of tea safety in the context of consumer preferences. In recent years, the advancement in sensor technology has replaced the human olfaction system with an artificial olfaction system, i.e., electronic noses (E-noses) for quality control of teas to differentiate the distinct aromas. Therefore, in this review, the potential applications of E-nose as a monitoring device for different teas have been investigated. The instrumentation, working principles, and different gas sensor types employed for E-nose applications have been introduced. The widely used statistical and intelligent pattern recognition methods, namely, PCA, LDA, PLS-DA, KNN, ANN, CNN, SVM, etc., have been discussed in detail. The challenges and the future trends for E-nose devices have also been highlighted. Overall, this review provides the insight that E-nose combined with an appropriate pattern recognition method is a powerful non-destructive tool for monitoring tea quality. In future, E-noses will undoubtedly reduce their shortcomings with improved detection accuracy and consistency by employing food quality testing.

Identification of railroad ballast fouling through statistical process control on ballast particle movement

Ballast fouling is an undesirable condition which adversely impacts track stability. As early detection of this condition is critical for the safety of train operations, various inspection technologies have been developed to provide information about state of the ballast and its substructure. Several studies have been conducted with the aim of providing a method for automatic and continuous monitoring of the track system. Ongoing efforts are focused on the development of techniques that enable real-time analysis of the ballast condition while remaining compatible with existing technologies. As an attempt toward that end, this study was designed to utilize statistical pattern recognition technique with data input from an innovative wireless sensor called RTS SmartRockTM (referred as SmartRock in this paper). This method was implemented in a field experiment on two sections of railroad track with homogenous traffic but different track conditions: one with clean ballast and the other with mud pumping. Four SmartRock sensors were installed at different locations on each section. With time-history data from the SmartRocks, an unsupervised statistical model was developed and evaluated for damage detection. The algorithm used was successful in distinguishing between data sets obtained from different ballast conditions. The results of the study show potential suitability of this method for timely detection of ballast fouling, thereby facilitating maintenance planning of rail tracks for a safer and more efficient transportation system.

An MPPCA-based approach for anomaly detection of structures under multiple operational conditions and missing data

Structural anomaly detection based on the structural health monitoring (SHM) data has attracted significant attention owing to its important role in the early warning of structural damage to existing civil structures. Data-driven approaches, where damage-sensitive features are extracted directly from the SHM data using statistical pattern recognition (SPR) techniques without physical models of structures, have been widely studied. Principal component analysis (PCA) and probabilistic PCA (PPCA) are powerful and efficient SPR methods for linear or weakly nonlinear cases. However, some special structures may be subjected to multiple operational conditions, wherein structural configurations such as geometry and mass distribution may change due to the movement of parts or the whole structure, as in retractable roof structures. These changes may give erroneous results in the SPR of the SHM data and eventually in the anomaly detection by a single PCA or PPCA model. This paper presents an improved approach using a mixture of probabilistic principal component analysis (MPPCA) for the anomaly detection of structures under multiple operational conditions with missing measurement data. First, the baseline MPPCA model was constructed for stress data collected under healthy conditions, where the estimation of the MPPCA parameters was reformulated for the missing data cases. Second, three anomaly statistics were presented for newly monitored incomplete data to detect and localize structural anomalies. The probability distributions of the anomaly statistics were estimated to obtain thresholds for outlier detection. Finally, the effectiveness of the MPPCA-based method was investigated by applying the method to the anomaly detection of a retractable roof structure with numerically simulated and real monitored stress data.

An investigation of rail failure due to wear using statistical pattern recognition techniques

The purpose of controlling and detecting errors in transportation systems is to ensure the correct operation of the system. The purpose of this paper is to identify wear in rails by using statistical pattern recognition methods. Due to the low accuracy of some of the current methods designing and introducing an easy way to be used in this area is required. The data of this paper were obtained by the field observation and using image captions of without worn profile and worn profile of railway rails, then the wear phenomenon was identified using the image processing method. Comparing the results of wear identification using pattern recognition techniques is in accordance with field observations. The true positive rate (sensitivity) is calculated and the accuracy of the proposed method indicates the validity of the proposed model.

Voice Pathologies Classification Using GMM And SVM Classifiers

In this paper we investigate the proprieties of automatic speaker recognition (ASR) to develop a system for voice pathologies detection, where the model does not correspond to a speaker but it corresponds to group of patients who shares the same diagnostic. One of essential part in this topic is the database (described later), the samples voices (healthy and pathological) are chosen from a German database which contains many diseases, spasmodic dysphonia is proposed for this study. This problematic can be solved by statistical pattern recognition techniques where we have proposed the mel frequency cepstral coefficients (MFCC) to be modeled first, with gaussian mixture model (GMM) massively used in ASR then, they are modeled with support vector machine (SVM). The obtained results are compared in order to evaluate the more preferment classifier. The performance of each method is evaluated in a term of the accuracy, sensitivity, specificity. The best performance is obtained with 12 coefficientsMFCC, energy and second derivate along SVM with a polynomial kernel function, the classification rate is 90% for normal class and 93% for pathological class.This work is developed under MATLAB

Bias correction for linear discriminant analysis

• We present an asymptotically exact estimator of the optimal intercept for the linear discriminant analysis (LDA). • The proposed accurate estimator is developed based on the theory of random matrices of increasing dimension. • The simple form of the estimator provides analytical insights into the mechanism of the proposed bias correction in LDA. • Analytical insights are complemented by empirical observations. Linear discriminant analysis (LDA) is perhaps one of the most fundamental statistical pattern recognition techniques. In this work, we explicitly present, for the first time, an asymptotically exact estimator of the LDA optimal intercept in terms of achieving the lowest overall risk in the classification of two multivariate Gaussian distributions with a common covariance matrix and arbitrary misclassification costs. The proposed estimator of the optimal bias term is developed based on the theory of random matrices of increasing dimension in which the observation dimension and the sample size tend to infinity while keeping their magnitudes comparable. The simple form of this estimator provides us with some analytical insights into the working mechanism of the bias correction in LDA. We then complement these analytical insights with numerical experiments. In particular, empirical results using real data show that insofar as the overall risk is concerned, the proposed bias-corrected form of LDA can outperform the conventional LDA classifier in a wide range of misclassification costs. At the same time, the superiority of the proposed form over LDA tends to be more evident as dimensionality or the ratio between class-specific costs increase.

Automated extraction of origin-destination demand for public transportation from smartcard data with pattern recognition

Origin-destination travel demand matrix is the signature of travel dynamics in transportation networks. Many fundamental analyses of transportation systems rely on the origin–destination demand matrix of the network. Although extraction of origin–destination travel demand for public transportation networks from ticketing data is not a new problem, yet it entails challenges, such as ‘alighting transaction inference’ and ‘transfer identification’ which are worthy of further attention. This is mainly because the state-of-the-art solutions to these challenges, are often heavily reliant on network-specific expert knowledge and extensive parameter setting, or multiple data sources. In this paper, we propose a procedure that effectively applies statistical pattern recognition techniques to address the main challenges in extracting the origin–destination demand from passenger smartcard records. Learning from patterns in the available data allows the procedure to perform well under minimum case-specific assumptions, thus it becomes applicable to smartcard data from various public transportation systems. The performance of the proposed framework is tested on a dataset of over 100 million smartcard transaction records from Melbourne’s multi-modal public transportation network. Evaluations on different aspects of the proposed procedure, suggest that the identified tasks are well addressed, and the framework is able to extract an accurate estimation of the origin–destination demand matrix for the system.

A Koopman operator approach for machinery health monitoring and prediction with noisy and low-dimensional industrial time series

Data-driven methods for machinery health monitoring and prediction, such as machine learning and statistical pattern recognition techniques, normally requires high quality (less noised), frequently-sampled and large volume time-series data to estimate proper models between system inputs and outputs. However, most of the industrial time-series are highly noisy and only few types of sensory data are available, which challenges the estimation accuracy. This paper proposes a data-driven spectral decomposition framework (denoted as Koopman-CBM) for the machinery health monitoring and prediction problem. Specifically, considering noisy industrial signals in the form of one dimensional time-series, we use the higher-order dynamic mode decomposition (DMD) embeds time-lagged snapshots to increase the spatial complexity of low-dimensional time series and use the total-least-square algorithm to compensate the effect of measurement noise, thereby, extracting accurate dynamical features. The obtained de-noised model characteristics (i.e., eigenvalues, eigenfunctions, Koopman operator) is effective in predicting the system health stages. In parallel, using the Koopman operator as the linear predictor associated with the nonlinear dynamics, we can then perform remaining useful life (RUL) predictions with high accuracy. The experimental validation of the proposed framework is carried out on a rolling bearing datasets for degradation health-stage inspection and RUL prediction. Results show that– compared with other mainstream methods– our approach is capable of identifying the critical degradation stages and achieving higher RUL prediction accuracies.

From waste to food: Optimising the breakdown of oil palm waste to provide substrate for insects farmed as animal feed.

Waste biomass from the palm oil industry is currently burned as a means of disposal and solutions are required to reduce the environmental impact. Whilst some waste biomass can be recycled to provide green energy such as biogas, this investigation aimed to optimise experimental conditions for recycling palm waste into substrate for insects, farmed as a sustainable high-protein animal feed. NMR spectroscopy and LC-HRMS were used to analyse the composition of palm empty fruit bunches (EFB) under experimental conditions optimised to produce nutritious substrate rather than biogas. Statistical pattern recognition techniques were used to investigate differences in composition for various combinations of pre-processing and anaerobic digestion (AD) methods. Pre-processing methods included steaming, pressure cooking, composting, microwaving, and breaking down the EFB using ionic liquids. AD conditions which were modified in combination with pre-processing methods were ratios of EFB:digestate and pH. Results show that the selection of pre-processing method affects the breakdown of the palm waste and subsequently the substrate composition and biogas production. Although large-scale insect feeding trials will be required to determine nutritional content, we found that conditions can be optimised to recycle palm waste for the production of substrate for insect rearing. Pre-processing EFB using ionic liquid before AD at pH6 with a 2:1 digestate:EFB ratio were found to be the best combination of experimental conditions.

Structural Health Monitoring of a Prestressed Concrete Bridge Based on Statistical Pattern Recognition of Continuous Dynamic Measurements Over 14 Years.

This work describes a vibration-based structural health monitoring of a prestressed-concrete box girder bridge on the A100 Highway in Berlin by applying statistical pattern recognition technique to a huge amount of data continuously collected by an integrated monitoring system during the period from 2000 to 2013. Firstly, the general condition and potential damage of the bridge is described. Then, the dynamic properties are extracted from 20 velocity sensors. Environmental variability captured by five thermal transducers and traffic intensity approximately estimated by strain measurements are also reported. Nonlinear influences of temperature on natural frequencies are observed. Subsequently, the measurements during the first year are used to build a baseline health index. The multiple linear regression (MLR) method is used to characterize the nonlinear relationship between natural frequencies and temperatures. The Euclidean distance of the residual errors is calculated to build a statistical health index. Finally, the indices extracted from the following years gradually deviate; which may indicate structural deterioration due to loss of prestress in the prestressed tendons.

Statistical Pattern Recognition based Technique for PV Array Fault Diagnosis

A reliable operation of photovoltaic (PV) systems should be guaranteed by implementing an effective fault detection program so that the maintenance schedule can be arranged to repair or replace the faulty component in time. This paper presents a PV array fault diagnosis method by applying the statistical pattern recognition technique. Proper feature parameters are selected first to characterize the operation condition of a PV array. The fault data set is then collected under varying operating conditions. Once the fault categories are defined, the genetic algorithm-based fuzzy C-means (GAFCM) clustering algorithm is employed to obtain the clustering center of each fault category so that the fuzzy mapping relationship between chosen feature parameters and its corresponding fault type is established. In this way, the fault type of a PV array can be judged immediately once the values of characteristic parameters are given. Finally, a Gaussian distribution-based membership function is developed to calculate the similarity between test samples and the defined fault categories. The fault type of the test sample can be recognized by judging from the largest similarity.

Data-driven damage diagnosis under environmental and operational variability by novel statistical pattern recognition methods

Feature extraction by time-series analysis and decision making through distance-based methods are powerful and efficient statistical pattern recognition techniques for data-driven structural health monitoring. The motivation of this article is to propose an innovative residual-based feature extraction approach based on AutoRegressive modeling and a novel statistical distance method named as Partition-based Kullback–Leibler Divergence for damage detection and localization by using randomly high-dimensional damage-sensitive features under environmental and operational variability. The key novel element of the proposed feature extraction approach is to establish a two-stage offline and online learning algorithms for extracting the residuals of AutoRegressive model as the main damage-sensitive features. This technique brings the great benefit of reducing the computational time and storage space for feature extraction in long-term monitoring conditions. The major contribution of Partition-based Kullback–Leibler Divergence method is to exploit a partitioning strategy for dividing random features into individual partitions and utilize numerical information of partitioning in distance calculation rather than directly applying random samples. Dealing with the major challenging issue of using the high-dimensional features in decision making and applicability to both correlated and uncorrelated random datasets are the main advantages of Partition-based Kullback–Leibler Divergence method. The accuracy and reliability of the proposed approaches are experimentally validated by two well-known benchmark structures. The stationarity and linearity of measured vibration responses for using in AutoRegressive modeling are evaluated by two hypothesis tests. Comparative studies are also conducted to demonstrate the superiority of the proposed methods over some exciting state-of-the-art techniques. Results show that the methods presented here succeed in detecting and locating damage and make time-saving and efficient tools for feature extraction and damage diagnosis.

Performance Factors of Czech Companies Identified Using Statistical Pattern Recognition: Interpretation of Results

The paper interprets factors of corporate performance identified by means of statistical pattern recognition techniques. A Dependency-Aware Feature algorithm with non-linear regression model ranked 74 potential factors of corporate performance according to their contribution to corporate performance prediction. This paper brings consecutive statistical analyses, which interpret the effects of Strategy, FDI, Share of Export, Top Management Performance Pay, and Workers' Performance Pay on corporate performance. Furthermore, the analyses reveal strong mutual moderating interdependencies. On the national scale, the paper brings evidence that the companies from the industries researched can use the stational techniques to learn about corporate performance factors. On a global scale, the paper introduces the contribution of Dependency-Aware Feature selection in the field of management and confirms the need for a multidimensional contingency approach in researching corporate performance. The results are based on a sample of 432 private limited or joint stock companies located in the Czech Republic operating in the manufacturing and construction industries and employing 50 or more people.

Antipyretic Effect of Herba Ephedrae-Ramulus Cinnamomi Herb Pair on Yeast-Induced Pyrexia Rats: A Metabolomics Study.

To investigate the antipyretic mechanism of Herba Ephedrae (Eph)-Ramulus Cinnamomi (RC) herb pair on yeast-induced pyrexia in rats. Totally 30 qualified male SD rats were randomly assigned to the normal control (NC) group, the pyrexia model (model) group, the Eph, RC and Eph-RC treatment groups by a random digital table, 6 rats in each group. Each rat received a 20% aqueous suspension of yeast (10 mL/kg) except the NC group. The 3 treatment groups were administered 8.1, 5.4 and 13.5 g/kg Eph, RC and Eph-RC respectively at 5 and 12 h after yeast injection, the NC group and the model groups were administered equal volume of distilled water. Rectal temperatures were measured at 0, 6, 8, 10, 12, 15, 18, 24 and 30 h and urine was collected prior to yeast injection and at 6, 10, 18, 24, 30, and 36 h after yeast injection. Then urine metabolomic profiling by gas chromatography tandem mass spectrometry, coupled with multivariate statistical analysis and pattern recognition techniques were used to explore the antipyretic effects of Eph-RC. Partial least squares discriminate analysis was used to analyze the metabolomics dataset including classification and regression in metabolomics plot profiling. Compared with the NC group, rectal temperatures were significantly higher in the model group (P<0.01), while 3 treatment groups decreased significantly compared with the model group (P<0.05 or P<0.01). Rectal temperatures of Eph-RC-treated rats started to go down at 6 h, and markedly decreased at 8, 12, 15, 18 and 24 h (P<0.05 or P<0.01), while those of the Eph and RC groups had decreased firstly at 8 h and were markedly lower at 12 h (P<0.05 or P<0.01). Seventeen potential biomarkers related to pyrexia were confirmed and identified, including pyruvic acid, L-phenylalanine, L-tyrosine, phenylacetic acid, hippuric acid, succinic acid, citrate and so on. Eight potential alterations of metabolic pathways including phenylalanine metabolism, citrate cycle, tryptophan metabolism, biosynthesis of valine, leucine and isoleucine, were identified in relation to the antipyretic effects of Eph-RC using MetPA software. The antipyretic effect of Eph-RC herb pair on yeast-induced pyrexia in rats involved correction of perturbed amino acid, fatty acid, and carbohydrate metabolism according to the metabolic pathway analysis with MetPA.

Phytochemical profile of Orthosiphon aristatus extracts after storage: Rosmarinic acid and other caffeic acid derivatives

BackgroundOrthosiphon aristatus (Blume) Miq. is a medicinal herb which is traditionally used for the treatment of diabetes and kidney diseases in South East Asia. Previous studies reported higher concentration of antioxidative phytochemicals, especially rosmarinic acid (ester of caffeic acid) and other caffeic acid derivatives in this plant extract than the other herbs such as rosemary and sage which are usually used as raw materials to produce rosmarinic acid supplement in the market. PurposeThe phytochemical profile of O. aristatus was investigated at different storage durations for quality comparison. MethodsThe phytochemicals were extracted from the leaves and stems of O. aristatus using a reflux reactor. The extracts were examined for total phenolic and flavonoid contents, as well as their antioxidant capacities, in terms of radical scavenging, metal chelating and reducing power. The phytochemical profiles were also analyzed by unsupervised principal component analysis and hierarchical cluster analysis, in relation to the factor of storage at 4 °C for 5 weeks. ResultsThe leaf extract was likely to have more phytochemicals than stem extract, particularly caffeic acid derivatives including glycosylated and alkylated caffeic acids. This explains higher ratio of total phenolic content to total flavonoid content with higher antioxidant capacities for the leaf extracts. Rosmarinic acid dimer and salvianolic acid B appeared to be the major constituents, possibly contributing to the previously reported pharmacological properties. However, the phytochemical profiles were found changing, even though the extracts were stored in the refrigerator (4 °C). The change was significantly observed at the fifth week based on the statistical pattern recognition technique. ConclusionO. aristatus could be a promising source of rosmarinic acid and its dimer, as well as salvianolic acid B with remarkably antioxidant properties. The phytochemical profile was at least stable for a month stored at 4 °C. It is likely to be a good choice of herbal tea with comparable radical scavenging activity, but lower caffeine content than other tea samples.

A Gaussian process–based approach to cope with uncertainty in structural health monitoring

Structural health monitoring is widely applied in industrial sectors as it reduces costs associated with maintenance intervals and manual inspections of damage in sensitive structures, while enhancing their operation safety. A major concern and current challenge in developing “robust” structural health monitoring systems, however, is the impact of uncertainty in the input training parameters on the accuracy and reliability of predictions. The aim of this article is to adapt an advanced statistical pattern recognition technique capable of considering variations in input parameters and arriving at a new structural health monitoring system more immune to the effect of uncertainty. Gaussian processes have been implemented to predict the state of damage in a typical composite airfoil structure. Different covariance functions were evaluated during the training stage of structural health monitoring. Results through a case study showed a remarkable capability of the Gaussian process–based approach to deal with uncertainty in the pattern recognition problem in structural health monitoring of a multi-layer composite airfoil structure. To illustrate robustness advantage of the approach as compared to conventional neural network models, the damage size and location prediction accuracy of the Gaussian process structural health monitoring has been compared to multi-layer perceptron neural networks. Some practical insights and limitations of the approach have also been outlined.

A damage assessment technique based on a revised Statistical Pattern-recognition Technique (SPRT)

Structural conditions assessment utilizes data analysis methods based on signals measured from subject structures. Among the various signal-based analysis methods available, a Statistical Pattern-recognition Technique (SPRT) based on Mahalanobis Distance (MD) theory is considered useful in assessing structural conditions; however, its evaluation of volatile signal data incurred by external loads such as seismic loads is usually limited. In this study, an experiment was conducted to attempt to overcome such limitations of MD theory. Furthermore, research was conducted on condition assessment methods based on SPRTs that can accurately assess structural conditions, even with highly volatile measured data. As a first step, the limitations of the standard MD theory were experimentally confirmed. Based on these limitations, a revised MD (RMD) theory was subsequently proposed. To verify the proposed SPRT, standardized cyclic loads and non-standardized seismic loads were used to conduct experiments on structural damage assessment. Based on these results, the RMD theory proposed in this study was confirmed to be a valid method for comprehending structural damage.

Use of Comprehensive Two-Dimensional Gas Chromatography with Time-of-Flight Mass Spectrometric Detection and Random Forest Pattern Recognition Techniques for Classifying Chemical Threat Agents and Detecting Chemical Attribution Signatures.

In this proof of concept study, chemical threat agent (CTA) samples were classified to their sources with accuracies of 87-100% by applying a random forest statistical pattern recognition technique to analytical data acquired by comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometric detection (GC × GC-TOFMS). Three organophosphate pesticides, chlorpyrifos, dichlorvos, and dicrotophos, were used as the model CTAs, with data collected for 4-6 sources per CTA and 7-10 replicate analyses per source. The analytical data were also evaluated to determine tentatively identified chemical attribution signatures for the CTAs by comparing samples from different sources according to either the presence/absence of peaks or the relative responses of peaks. These results demonstrate that GC × GC-TOFMS analysis in combination with a random forest technique can be useful in sample classification and signature identification for pesticides. Furthermore, the results suggest that this combination of analytical chemistry and statistical approaches can be applied to forensic analysis of other chemicals for similar purposes.