Principal Component Analysis Model Research Articles

Мultivariate analysis for rapid screening and prediction of solid-state compatibility in pharmaceutical preformulation studies-paving the road for machine learning

Multivariate analysis models were developed to evaluate the results obtained from a compatibility study designed for ibuprofen with a large group of different types of excipients, as a possible approach for rapid screening of the incompatibility between the active pharmaceutical ingredient (API) and excipients. The solid-state characterization of the binary mixtures was performed by Fourier transform infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC). Principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) using SIMCA® software were applied for evaluation of the experimentally obtained results. The optimal PCA model for the FTIR spectra explains 96.2 % of the variations in the dataset with good statistical indicators (R2X = 0.960, Q2 = 0.900), which was also the case for the PCA model for the DSC curves (R2X = 0.981, Q2 = 0.866). The applied PLS-DA models have shown similar behaviour to the PCA. Moreover, the main spectral variations in the FTIR spectra and the thermal events in the DSC data were attributed the highest variable importance for the projection (VIP) scores in the corresponding VIP plots, confirming the model capability for predicting ibuprofen interactions. Furthermore, the prediction power of the optimal models for FTIR and DSC experimental data was evaluated by the root mean square error of prediction (RMSEP) of 0.10 and 0.16, respectively. The obtained results demonstrated the potential of multivariate statistical analysis as a machine learning-based technique for screening and prediction of ibuprofen-excipients solid-state compatibility in the preformulation phase of the pharmaceutical development of dosage forms.

Ecological risk and spatial distribution, sources of heavy metals in typical purple soils, southwest China

The identification and quantification of the ecological risks, sources and distribution of heavy metals in purple soils are essential for regional pollution control and management. In this study, geo-accumulation index (Igeo), enrichment factor (EF), pollution index (PI), potential ecological risk index (RI), principal component analysis (PCA) model and geographical detector (GD) were combined to evaluate the status, ecological risk, and sources of heavy metals (HMs) in soils from a typical purple soil areas of Sichuan province. The results showed that the average contents of As, Cd, Cr, Cu, Hg, Ni, Pb and Zn in purple soil were 7.77, 0.19, 69.5, 27.9, 0.077, 30.9, 26.5 mg/kg and 76.8 mg/kg, and the Igeo, EF and RI of topsoil Hg and Cd in designated area was the highest, and the average contents of Hg and Cd in topsoil were obviously greater than respective soil background value in Sichuan province and purple soil. The hot spots for the spatial distribution of 8 HMs were mainly focused in the southwest and northeast of the designated area, and there were also significant differences for 8 HMs distribution characteristics in the profile soil. Cu comes from both anthropogenic and natural sources, Zn, Ni and Cr mainly come from natural sources, but As, Pb, Hg and Cd mainly derived from human activities. GD results showed that soil texture (X18), altitude (X4), total nitrogen (TN), clay content (X3), sand content (X2) and silt content (X1) had the greatest explanatory power to 8 HMs spatial differentiation.This study provides a reference for understanding the status and influencing factors of HM pollution in typical purple soil, and lays a theoretical foundation for the environmental treatment of purple soil in China.

Assessment of water quality using entropy-weighted quality index, statistical methods and electrical resistivity tomography, Moti village, northern Pakistan

In this study, twenty-two water samples were collected from boreholes (BH), and streams to evaluate drinking water quality, its distribution, identification of contamination sources and apportionment for Moti village, northern Pakistan. An atomic absorption spectrophotometer (AAS) is utilized to determine the level of heavy metals in water such as arsenic (As), zinc (Zn), lead (Pb), copper (Cu), cadmium (Cd), manganese (Mn), and ferrous (Fe). Groundwater chemistry and its quantitative driving factors were further explored using multivariate statistical methods, Principal Component Analysis (PCA) and Positive Matrix Factorization (PMF) models. Finally, a total of eight electrical resistivity tomographs (ERTs) were acquired across i) the highly contaminated streams; ii) the villages far away from contaminated streams; and iii) across the freshwater stream. In the Moti village, the mean levels (mg/l) of heavy metals in water samples were 7.2465 (As), 0.4971 (Zn), 0.5056 (Pb), 0.0422 (Cu), 0.0279 (Cd), 0.1579 (Mn), and 0.9253 (Fe) that exceeded the permissible limit for drinking water (such as 0.010 for As and Pb, 3.0 for Zn, 0.003 for Cd and 0.3 for Fe) established by the World Health Organization (WHO, 2018). The average entropy weighted water quality index (EWQI) of 200, heavy metal pollution index (HPI) of 175, heavy metal evaluation index (HEI) of 1.6 values reveal inferior water quality in the study area. Human health risk assessment, consisting of hazard quotient (HQ) and hazard index (HI), exceeded the risk threshold (>1),indicating prevention of groundwater usage. Results obtained from the PCA and PMF models indicated anthropogenic sources (i.e. industrial and solid waste) responsible for the high concentration of heavy metals in the surface and groundwater. The ERTs imaged the subsurface down to about 40 m depths and show the least resistivity values (<11 Ωm) for subsurface layers that are highly contaminated. However, the ERTs revealed relatively high resistivity values for subsurface layers containing fresh or less contaminated water. Filtering and continuous monitoring of the quality of drinking water in the village are highly recommended.

An effective ultrasound features-based diagnostic model via principal component analysis facilitated differentiating subtypes of mucinous breast cancer from fibroadenomas

BackgroundMucinous breast carcinoma (MBC) risks misdiagnosis as a fibroadenoma (FA), leading to inappropriate or delayed treatments. This study aimed to establish an efficient ultrasound (US)-based diagnostic model for distinguishing MBC subtypes from FAs MethodsBetween January 2017 to February 2024, 240 lesions were enrolled, comprising 65 cases of pure mucinous breast carcinoma (PMBC), 47 cases of mixed mucinous breast carcinoma (MMBC), and 128 cases of FAs. Ten US feature variables underwent principal component analysis (PCA), models were constructed based on components explaining over 75% of the total variation, and varimax rotation was applied. Then, comprehensive models were constructed for distinguishing PMBCs or MMBCs from FAs ResultsSix principal components were selected, achieving a cumulative contribution rate of 77.46% (PMBCs vs. FAs) and 78.62% (MMBCs vs. FAs). The principal component of cystic-solid composition and posterior acoustic enhancement demonstrated the highest diagnostic value (AUC:0.86, ACC: 80.31%) for distinguishing PMBCs from FAs. Features including vascularization, irregular shape, ill-defined border, and larger size exhibited the highest diagnostic value (AUC:0.90, ACC:87.43%) for distinguishing MMBCs from FAs. The comprehensive models showed excellent clinical value in distinguishing PMBCs (AUC = 0.86, SEN = 86.15%, SPE = 73.44%, ACC = 77.72%) and MMBCs (AUC = 0.92, SEN = 80.85%, SPE = 95.31%, ACC = 91.43%) from FAs ConclusionThis diagnostic model holds promise for effectively distinguishing PMBCs and MMBCs from FAs, assisting radiologists in mitigating diagnostic biases and enhancing diagnostic efficiency. MicroAbstractDistinguishing between Mucinous Breast Carcinoma (MBC) and Fibroadenoma (FA) by ultrasound presents a challenge. This study collected 240 cases of MBC and FA and developed principal component analysis models based on ultrasound features to differentiate pure MBC and mixed MBC from FA. Our findings revealed a high diagnostic efficacy for this approach, demonstrating its potential in clinical practice.

Non-Destructive Prediction of the Mixed Mineral Pigment Content of Ancient Chinese Wall Paintings Based on Multiple Spectroscopic Techniques.

This study first developed non-destructive and accurate methods to predict the relative contents of mixed mineral pigments in ancient Chinese wall paintings using multiple spectroscopic techniques. The colorimetry, attenuated total reflection Fourier transform infrared spectroscopy (ATR FT-IR), ultraviolet-visible-near-infrared (UV-Vis-NIR) spectroscopy, and Raman spectroscopy were employed. Analyses were conducted including color difference, spectral reflection, ATR FT-IR spectra, and Raman mapping for simulated samples (malachite-lazurite mixed with rabbit glue samples) before and after aging. Models were then established for predicting the relative pigment contents of samples using UV-Vis-NIR and ATR FT-IR spectral data with Beer-Lambert law, and mathematical methods comprising principal component analysis (PCA) and nonlinear curve fitting. In particular, PCA and empty modeling methods combined with non-negative partial least squares were developed to predict the relative pigment contents based on Raman mapping data. The results demonstrated that approaches comprising PCA, mathematical model, and empty modeling based on the spectral data were effective at predicting the relative pigment contents. The predicted results obtained using the mathematical model based on UV-Vis-NIR spectra had an error of about 2%, and the best prediction based on ATR FT-IR spectra had an error of <3.6% at 1041 cm-1. The errors for the predictions using PCA and empty modeling based on Raman mapping data were 0.01-9.30% and 0.28-7.15%, respectively. However, the predicted relative pigment contents obtained based on ATR FT-IR data combined with the Beer-Lambert law had higher errors. The findings of this study confirm the strong feasibility of using spectroscopic techniques for quantitatively analyzing mixed mineral pigments.

Assessing antecedents of Google shopping ads intention to purchase: a multigroup analysis of generation Y and Z

Purpose The substantial expansion of technology and the efficacy of digital platforms in reaching young audiences have led to enhanced targeting and customization of promotional communications. Notwithstanding the expansion and efficacy of contemporary advertising platforms, scholarly attention has not kept pace with this domain of inquiry. This study aims to assess the antecedents of Google Shopping Ads (GSA) on intention to purchase behavior among the Generation Y and Z cohorts. Design/methodology/approach The current study used a quantitative approach and snowball sampling technique to gather primary data via a questionnaire and Google Forms, which resulted in the collection of 5,808 questionnaires among the cohort members. A principal component analysis and multigroup confirmatory multigroup structural equation modeling (between Generation Y and Z) were used to assess the research data and model. Findings The results show positive trust and perceived value associations with intention to purchase, particularly among Generation Y and Z consumers. The findings also show negative irritation, product risk and time risk associations with intention to purchase, especially among the Generation Y cohort, which indicates that young consumers generally do not observe perceived risk due to the usage of GSA. Originality/value GSA will continue to grow and become an increasingly important integrated marketing communications tool as the digital landscape develops. It can be concluded that young consumers show a high degree of perceived value and low levels of perceived risk due to the use of GSA. This study, therefore, promotes improved understanding among academics, marketers and businesses of search engine advertising among young cohorts of consumers (Generation Y and Z) in a developing country context.

Pride (in the name of tourism): Mitigating the effects of overtourism on festivals

During popular festivals, overtourism can be a significant problem that negatively affects many tourist destinations. This study carried out a systematic review of the literature on pride and the effects of perceived overtourism. Based on the variables and concepts found in the literature, a questionnaire was developed with 37 items that reflect different dimensions and stages of mitigating the effects of overtourism. Experts were consulted for questionnaire validation, and the validation process was conducted through a sequential procedure. The questionnaire's psychometric properties were tested using principal component analysis (PCA) and least squares partial structural equation modeling (PLS-SEM). The results indicated that the developed scale has good internal consistency and construct validity. The results demonstrate that cultural and economic aspects strongly influence residents' pride, giving them a more positive perception of the festival, despite the challenges associated with tourist saturation.

Differentiating Structurally Similar Fentanyl Analogs by Comparing Density Functional Theory (DFT) Calculations and Surface-Enhanced Raman Spectroscopy (SERS) Results.

Fentanyl and fentanyl analogs are the main cause of recent overdose deaths in the United States. The presence of fentanyl analogs in illicit drugs makes it difficult to estimate their potencies. This makes the detection and differentiation of fentanyl analogs critically significant. Surface-enhanced Raman spectroscopy (SERS) can differentiate structurally similar fentanyl analogs by yielding spectroscopic fingerprints for the detected molecules. In previous years, five fentanyl analogs, carfentanil, furanyl fentanyl, acetyl fentanyl, 4-fluoroisobutyryl fentanyl (4-FIBF), and cyclopropyl fentanyl (CPrF), gained popularity and were found in 76.4% of the fentanyl analogs trafficked. In this study, we focused on 4-FIBF, CPrF, and structurally similar fentanyl analogs. We developed methods to differentiate these fentanyl analogs using theoretical and experimental methods. To do this, a set of fentanyl analogs were examined using density functional theory (DFT) calculations. The DFT results obtained in this project permitted the assignment of spectral bands. These results were then compared with normal Raman and SERS techniques. Structurally similar fentanyl analogs show important differences in their spectra, and they have been visually differentiated from each other both theoretically and experimentally. Additional results using principal component analysis and soft independent modeling of class analogy show they can be distinguished using this technique. The limit of detection values for FIBF and CPrF were determined to be 0.35 ng/mL and 4.4 ng/mL, respectively, using SERS. Experimental results obtained in this project can be readily implemented in field applications and smaller laboratories, where inexpensive portable Raman spectrometers are often present and used in drug analysis.

HS-AFM single-molecule structural biology uncovers basis of transporter wanderlust kinetics.

The Pyrococcus horikoshii amino acid transporter GltPh revealed, like other channels and transporters, activity mode switching, previously termed wanderlust kinetics. Unfortunately, to date, the basis of these activity fluctuations is not understood, probably due to a lack of experimental tools that directly access the structural features of transporters related to their instantaneous activity. Here, we take advantage of high-speed atomic force microscopy, unique in providing simultaneous structural and temporal resolution, to uncover the basis of kinetic mode switching in proteins. We developed membrane extension membrane protein reconstitution that allows the analysis of isolated molecules. Together with localization atomic force microscopy, principal component analysis and hidden Markov modeling, we could associate structural states to a functional timeline, allowing six structures to be solved from a single molecule, and an inward-facing state, IFSopen-1, to be determined as a kinetic dead-end in the conformational landscape. The approaches presented on GltPh are generally applicable and open possibilities for time-resolved dynamic single-molecule structural biology.

The science behind the nose: correlating volatile organic compound characterisation with canine biodetection of COVID-19.

The SARS-CoV-2 pandemic stimulated the advancement and research in the field of canine scent detection of COVID-19 and volatile organic compound (VOC) breath sampling. It remains unclear which VOCs are associated with positive canine alerts. This study aimed to confirm that the training aids used for COVID-19 canine scent detection were indeed releasing discriminant COVID-19 VOCs detectable and identifiable by gas chromatography (GC-MS). Inexperienced dogs (two Labradors and one English Springer Spaniel) were trained over 19 weeks to discriminate between COVID-19 infected and uninfected individuals and then independently validated. Getxent tubes, impregnated with the odours from clinical gargle samples, used during the canines' maintenance training process were also analysed using GC-MS. Three dogs were successfully trained to detect COVID-19. A principal components analysis model was created and confirmed the ability to discriminate between VOCs from positive and negative COVID-19 Getxent tubes with a sensitivity of 78% and a specificity of 77%. Two VOCs were found to be very predictive of positive COVID-19 cases. When comparing the dogs with GC-MS, F1 and Matthew's correlation coefficient, correlation scores of 0.69 and 0.37 were observed, respectively, demonstrating good concordance between the two methods. This study provides analytical confirmation that canine training aids can be safely and reliably produced with good discrimination between positive samples and negative controls. It is also a further step towards better understanding of canine odour discrimination of COVID-19 as the scent of interest and defining what VOC elements the canines interpret as "essential".

Assessing the Impact of Nitrogen Fertilization, Variety Selection, Year and Their Interaction on Wheat Yield and Yield Components

This five-year study (2016–2021) in Morocco’s Mediterranean climate investigated the effect of nitrogen fertilization and genotypic selection on wheat yield and associated components. Utilizing a split-plot design, the study assessed five wheat genotypes—’Faraj’, ‘Luiza’, ‘Itri’, ‘Karim’ and ‘Nassira’—under three nitrogen application rates (120, 60 and 0 kg/ha) across thirty plots with two replicates. Interactions between nitrogen and year showed marked significance in yield (p = 0.001), biomass (p = 0.002), TKW (p = 0.003) and Spk/m2 (p = 0.001), underscoring the variability in optimal nitrogen application rates across different years. Additionally, significant interactions between variety and year were observed for biomass (p = 0.001) and G/m2 (p = 0.001), indicating variability in the performance of different varieties across years. The ‘Itri’ genotype showed the highest yield in 2017, while ‘Luiza’ was pre-eminent in 2018, with ‘Itri’ producing the most biomass. ‘Faraj’ demonstrated consistent superiority in yield and biomass during 2019 and 2020. Our integrated principal component analysis and quadratic models elucidated that an intermediate nitrogen rate of 60 kg/ha (N2) was particularly advantageous for the ‘Faraj’ and ‘Karim’ genotypes. These findings highlight the substantial impact of informed nitrogen level adjustment and genotypic selection on yield optimization.

A framework for optimizing environmental covariates to support model interpretability in digital soil mapping

A common practice in digital soil mapping (DSM) is to incorporate many environmental covariates into a machine-learning algorithm to predict the spatial patterns of soil attributes. Variance inflation factor (VIF), principal component analysis (PCA), and recursive feature elimination (RFE) are three statistical methods that can be used to reduce the number of covariates. This study aims 1) to compare VIF and PCA approaches; 2) to identify an approach to determine the minimum number of covariates in DSM to ensure model parsimony using RFE after using VIF; and 3) to examine methods to interpret the impact of covariates on the variability of the predicted soil properties. The study area was the province of British Columbia (BC), Canada. This study used legacy data for four soil properties to make digital soil maps: soil organic carbon (SOC%), pH, clay%, and coarse fragment (CF%). Seven models were made for each soil property to determine the influence on validation results by using a different number of covariates produced by various methods on validation results. The results showed that the number of covariates could be reduced from 70 to 4 to 12 with only a little or no difference in concordance correlation coefficient (CCC) validation results. The CCC results of pH models using 70 and 7 covariates were both 0.74, and for other soil properties, this difference was negligible. The validation results obtained from PCA models showed that the performance of PCA in reducing the number of covariates was not as effective as when using VIF. Moreover, this study showed that covariates related to precipitation were the most important for modeling SOC%, soil pH, and clay%. Topographic covariates were the most influential covariates for modeling soil CF%. This study emphasizes the potential benefits of combining various data reduction methods to achieve optimal outcomes and generate the most parsimonious and interpretable models.

Longitudinal Monitoring of Biomechanical and Psychological State in Collegiate Female Basketball Athletes Using Principal Component Analysis.

The growth in participation in collegiate athletics has been accompanied by increased sport-related injuries. The complex and multifactorial nature of sports injuries highlights the importance of monitoring athletes prospectively using a novel and integrated biopsychosocial approach, as opposed to contemporary practices that silo these facets of health. Data collected over two competitive basketball seasons were used in a principal component analysis (PCA) model with the following objectives: (i) investigate whether biomechanical PCs (i.e., on-court and countermovement jump (CMJ) metrics) were correlated with psychological state across a season and (ii) explore whether subject-specific significant fluctuations could be detected using minimum detectable change statistics. Weekly CMJ (force plates) and on-court data (inertial measurement units), as well as psychological state (questionnaire) data, were collected on the female collegiate basketball team for two seasons. While some relationships (n = 2) were identified between biomechanical PCs and psychological state metrics, the magnitude of these associations was weak (r = |0.18-0.19|, p < 0.05), and no other overarching associations were identified at the group level. However, post-hoc case study analysis showed subject-specific relationships that highlight the potential utility of red-flagging meaningful fluctuations from normative biomechanical and psychological patterns. Overall, this work demonstrates the potential of advanced analytical modeling to characterize components of and detect statistically and clinically relevant fluctuations in student-athlete performance, health, and well-being and the need for more tailored and athlete-centered monitoring practices.

Does New Urbanization Support the Rural Inclusive Green Development under Domestic Circulation in China?

New urbanization is an endogenous driving force to enhance domestic circulation. Driving the development of rural industries with urbanization to achieve interactive symbiosis has become an important topic to promote the coordinated development of urban and rural green. Based on the panel data of 30 provinces in China from 2009 to 2021, this paper constructs an evaluation index system for new urbanization and rural inclusive green development, and uses principal component analysis and panel regression model to analyze the impact of new-type urbanization on inclusive green development in rural areas. The results of the study show the following: (1) Rural inclusive green development and new urbanization have been significantly improved during the study period, but there are significant regional differences. (2) The construction of the new urbanization significantly promotes rural inclusive green development, but there is significant spatial heterogeneity. This effect is more significant in the Eastern and Central regions. (3) Population urbanization, land urbanization, social urbanization, and environmental urbanization can effectively promote rural inclusive green development, but economic urbanization will have a negative impact on green development in the countryside during the study period. Therefore, it is necessary to further strengthen the leading role of central cities and urban agglomerations, to promote the countryside with the city and at the same time to combat environmental pollution and to create ecologically livable towns and villages. In addition, the government should strengthen top-level design, provide industrial support to backward areas, improve the spatial layout of urbanization, and promote the deepening of new urbanization.

Multirate Mixture Probability Principal Component Analysis for Process Monitoring in Multimode Processes

In the multirate sampling processes, the process data are usually collected from various operating conditions and display multimodal characteristics. To monitor these multirate multimode processes, a multirate mixture probability principal component analysis model is proposed for process modeling and fault detection. In this model, the local multirate models are built first for each mode and all of them are subsequently fused with the mixture modeling approach. Such model is able to deal with multirate data with various amount of sampling rates, contributing to a remarkable fault detection and mode identification performance by utilizing all the available measurements even if some variables are unobserved. Then the expectation <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$-$</tex-math> </inline-formula> maximum algorithm is utilized to estimate all the model parameters in the probabilistic framework and the corresponding monitoring method is also developed based on the constructed models. Finally, the effectiveness of the proposed method is demonstrated through a PRONTO benchmark and a real multimode ammonia synthesis process. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Motivated by the practical problem of ununiform sampling intervals in multimode processes, this paper proposes a novel multirate mixture probability principle component analysis model for processes modeling and monitoring. In this model, all the available observations with different sampling rates can be incorporated, which contributes greatly to capturing the multimodal characteristics within the industrial processes. Such ability is the key to realize multimode process monitoring, evaluation, fault diagnosis, and process optimization. In addition, although this paper only focuses on the continuous multirate data in industry, it is equally applicable to other forms of multirate data, such as images and videos.

Experimental data-driven model development for ESP failure diagnosis based on the principal component analysis

AbstractThe reliable diagnosis of electrical submersible pump (ESP) failure is a vital process for establishing of the optimal production strategies and achieving minimum development costs. Although traditional ammeter charts and nodal analysis are commonly used for ESP failure diagnosis, the techniques have limitations, as it requires manpower and is difficult to diagnose the failure in real-time. Therefore, in this study, ESP failure diagnosis was performed using the principal component analysis (PCA). First, 11 types of 9,955 pieces of data were acquired from a newly constructed ESP experimental system for 300 days. During the experimental period, ESP failure occurred twice with a significant drop in performance: first on day 112 and second on day 271. The PCA model was constructed with the 8,928 pieces of normal status data and tested with the 1,027 pieces of normal and failure status data. Three principal components were extracted from the measured data to identify the patterns of the normal and failure status. Based on the logistic regression method to analyze the efficiency of the PCA model, it was found out that the developed PCA model showed an accuracy of 93.3%. Therefore, the PCA model was found to be reliable and effective for the ESP failure diagnosis and performance analysis.

Publisher Correction: The Effectiveness of a Probabilistic Principal Component Analysis Model and Expectation Maximisation Algorithm in Treating Missing Daily Rainfall Data

Publisher Correction: The Effectiveness of a Probabilistic Principal Component Analysis Model and Expectation Maximisation Algorithm in Treating Missing Daily Rainfall Data

Differentiation of blue gel pens using visible spectroscopy coupled with chemometric techniques

Ink analysis is one of the most important and keydimension of forensic document examination (FDE). It is performed to compare, identify, characterize and discriminate different kinds of inks present on questioned document. It becomes key evidence in forensic document examination. In the present work, thirty-eight blue gel pens are discriminated using visible spectroscopy coupled with chemometric techniques. The recorded spectra were visually compared followed by chemometric techniques of principal component analysis (PCA) and linear discriminant analysis (LDA). Discriminating power for visual comparison, PCA and LDA are 58.18%, 100% and 100% respectively. Developed PCA model is validated and 100% accuracy, precision, sensitivity and specificity are achieved. Blue gel pen inks were 100% correctly differentiated on the basis of their brands using PCA followed by LDA. The utilized approach is fast and provides reproducible results. The results of present work demonstrate that potential utility of conventional and cost-effective visible spectroscopy can be improved by coupling it with these chemometric techniques. Major advantage of present research is the minimum damage caused to document. This study establishes a method which provides proof of concept discrimination of the gel pen samples. The methodology adopted in present work can be applied in the alteration cases in examination of suspected documents.

New methods for the identification of malachite pigments with varying particle sizes used in ancient Chinese murals by spectroscopic techniques

This study attempts to propose a simple, low-interventive and higher accuracy identification method for malachite pigments with varying particle sizes based on attenuated total reflection-Fourier transform infrared (ATR-FTIR), and evaluate the accuracy of ATR-FTIR and ultraviolet–visible near-infrared (UV-VIS-NIR) spectroscopy for identifying particle sizes. The micro-morphology, particle size, colour difference, and spectral characteristics of the malachite samples were determined using scanning electron microscopy (SEM), laser particle size analysis, colourimetry, UV-VIS-NIR and ATR-FTIR spectrometry. The principal component analysis (PCA) and mathematical models based on nonlinear curve fitting using UV-VIS-NIR and FTIR spectral data were developed and evaluated for the particle size identification of various malachite samples (pure malachite samples, fresh and aged malachite-rabbit glue samples). SEM morphological results showed that the malachite pigments with smaller particle sizes appear to have irregular flakes with a few loose agglomerated granules attached to the surface. The weight-specific surface area also increased gradually as the pigment particle size decreased. The colour difference results revealed that smaller malachite particle size gradually shifted toward the red–blue regions in CIE L*a*b* colour space. In addition, more intense reflectance and IR spectral absorbance peaks were observed in the case of the smaller particle size. The above proposed methods of particle size identification proved to be highly feasible and effective in predicting pigment particle size, which were confirmed by the PCA identification results and the mathematical model. Also, ATR-FTIR spectroscopy could provide more accurate predictions than UV-VIS-NIR spectroscopy, by keeping prediction errors below 5.5 μm.

Anomaly detection in multivariate time series of drilling data

Different clusters of abnormal activities often arise within same temporal domain of drilling operations. This contrasts with employing simplified scenarios, such as anomaly detection models for specific issues like stuck pipe or loss circulation. The dynamic nature of drilling environments demands a holistic framework that can adapt to evolving conditions and detect anomalies beyond predefined categories. There exists a need to evaluate the performance of various data-driven models and achieve a broader applicability of anomaly detection instead of focusing on a single abnormal activity type. This study presents a generalized framework designed to detect anomalous events in drilling operations. Three different unsupervised learning algorithms, principal component analysis (PCA), isolation forest (IF) and lstm-autoencoder (LSTM-AE), are used to determine abnormal patterns and irregularities in the drilling data. PCA is chosen for its interpretability and efficiency in handling high-dimensional data by reducing the dimensionality. IF is selected for its effectiveness in isolating anomalies, making it robust for scenarios where anomalies exhibit distinct patterns. LSTM-AE is employed due to its capability to capture temporal dependencies and nonlinear patterns in time series data. A multivariate time series of drilling data from an offshore well is collected to use in this work. The dataset comprises approximately six months of drilling operations, including drilling and non-drilling periods. The input variables are chosen from the common controllable parameters continuously monitored by the drilling crew. The pre-processed drilling data with input features is organized into groups based on each drilling section. These unlabeled data is split into train, validation and test sets. Cross-validation (CV) is applied into train and validation sets to prevent overfitting and tune the hyperparameters. The test data is used to evaluate the generalizability and robustness of each model. Evaluation metrics such as precision, recall, F1 score, and receiver operating characteristic (ROC) curve are used to select the best-performing model. The thresholds for anomaly identification are determined by analyzing the reconstruction errors from PCA and LSTM-AE models, as well as anomaly scores from the IF algorithm. Subsequently, evaluation metrics are calculated using these thresholds, assigning true and false predictions to each data point. These predictions are then compared with the ground truth data labels obtained from the daily reports. Experimental results show that the proposed approach can effectively predict anomalous events in drilling operations. It is observed that there is a causal link between irregularities in time series drilling data and the occurrence of anomalous events in drilling operations. Implementing the proposed approach in real-time monitoring systems may improve overall efficiency and minimize risks in drilling operations. The presented methodology can be tested with diverse drilling datasets to enhance the generalizability and reduce bias, leading to a more comprehensive and reliable assessment across various scenarios.