Search

This project aims to mitigate the critical decline in bee populations, essential for crop pollination and food security. With a shortage of taxonomic specialists to identify the vast array of bee species, the project's goal is to enhance the monitoring of population changes through an automated classification system. Utilizing a dataset of bee wing images, the project aims to develop a computational pipeline to identify species based on their unique wing vein patterns. This approach not only supports bee conservation efforts but also expands our understanding of complex geometric variations in nature, offering wider applications in biological research. This poster was presented at the UCSB Data Science Capstone showcase in 2024.

Differences in gait patterns of children with Duchenne muscular dystrophy (DMD) and typically developing (TD) peers are visible to the eye, but quantifications of those differences outside of the gait laboratory have been elusive. In this work, we measured vertical, mediolateral, and anteroposterior acceleration using a waist-worn iPhone accelerometer during ambulation across a typical range of velocities. Fifteen TD and fifteen DMD children from 3 to 16 years of age underwent eight walking/running activities, including five 25 m walk/run speed-calibration tests at a slow walk to running speeds (SC-L1 to SC-L5), a 6-min walk test (6MWT), a 100 m fast walk/jog/run (100MRW), and a free walk (FW). For clinical anchoring purposes, participants completed a Northstar Ambulatory Assessment (NSAA). We extracted temporospatial gait clinical features (CFs) and applied multiple machine learning (ML) approaches to differentiate between DMD and TD children using extracted temporospatial gait CFs and raw data. Extracted temporospatial gait CFs showed reduced step length and a greater mediolateral component of total power (TP) consistent with shorter strides and Trendelenberg-like gait commonly observed in DMD. ML approaches using temporospatial gait CFs and raw data varied in effectiveness at differentiating between DMD and TD controls at different speeds, with an accuracy of up to 100%. We demonstrate that by using ML with accelerometer data from a consumer-grade smartphone, we can capture DMD-associated gait characteristics in toddlers to teens.

Forestiera includes woody plants distributed from the United States to Ecuador, with a wide distribution in Mexico. However, basic knowledge about its biology and taxonomy is still limited. In the present study, a leaf morphometric analysis of the 14 Mexican Forestiera species, including F. neomexicana, was carried out to determine if leaf characters allow classifying the species and to evaluate if the inclusion of qualitative features in statistical analyses improves the fit of used methods. Four quantitative and four qualitative variables from herbarium specimens, were evaluated. A matrix was generated which was submitted to multivariate analysis, including Linear Discriminant Analysis and Random Forest. Results revealed that the species are correctly classified, with Random Forest being the method with the best classifying capacity. The most important variables to differentiate the species are the shape of the base, width and margin of the leaves. All species were classified as different entities including F. pubescens and F. neomexicana, currently considered synonyms, were classified by Random Forest as different entities. [ABSTRACT FROM AUTHOR]

Statistician C. R. Rao made many contributions to multivariate analysis over the span of his career. Some of his earliest contributions continue to be used and built upon almost 80 years later, while his more recent contributions spur new avenues of research. The present article discusses these contributions, how they helped shape multivariate analysis as we see it today, and what we may learn from reviewing his works. Topics include his extension of linear discriminant analysis, Rao's perimeter test, Rao's U statistic, his asymptotic expansion of Wilks' Λ statistic, canonical factor analysis, functional principal component analysis, redundancy analysis, canonical coordinates, and correspondence analysis. The examination of his works shows that interdisciplinary collaboration and the utilization of real datasets were crucial in almost all of Rao's impactful contributions. [ABSTRACT FROM AUTHOR]

The broad term 'leukemia' refers to different types of cancer related to blood cells. Detecting and identifying the specific type of leukemia continues to be a major challenge in the medical field. Diverse machine learning techniques can be vital in analyzing gene expression data from microarray experiments in cancer research related to leukemia. In particular, the Leukemia Gene Expression data from the Curated Microarray Database (CuMiDa) is used here. Microarrays can be challenging in determining expression patterns. In this work, we use Fisher's linear discriminant analysis, a popular technique for dimensionality reduction, together with a new feature selection approach to predict leukemia using microarray data. Our machine learning model is used to predict five types of leukemia including AML, PBSC CD34, Bone Marrow, and CD34 from the bone marrow. This is achieved by first rescaling the data features. We then use a feature selection technique to obtain the 25 most significant features from the dataset's 22,283 features, then further reduce the dimension to 5 features only, to reduce computational complexity. These features are then fed into a Fisher's linear discriminant module and a likelihood-based index for classification. The overall performance of our model was excellent. We examine the results using 2, 4, 5, 6, and 7 selected features. The best classification accuracies are 89.6%, 96.92%, and 96.15%, for 2, 5, and 7 selected features, respectively. Our results outperform the state-of-the-art by about 4%, with an excellent task completion time of less than 100 ms. [ABSTRACT FROM AUTHOR]

This research paper introduces an innovative approach to classify heart rate variability (HRV) time series into paced and spontaneous breathing patterns to reflect changes in the autonomic nervous system. This type of classification is beneficial in wearable devices for stress/relaxation level detection and in deciding therapeutic interventions. The "Multi-Domain Approach" methodology integrates three different techniques: standard HRV features, fuzzy recurrence plot (FRP)-based FRP_GLCM, and empirical mode decomposition-based IMF_FRP_GLCM. The study concentrates on analyzing HRV time series within shorter data segments, aligning with the requirements of contemporary wearable health devices and biofeedback systems. HRV data collected during spontaneous and slow-paced breathing were analyzed across data segments of 5, 4, 3, 2, and 1 min, incorporating feature selection and reduction methods. Results demonstrated that standard HRV features yielded optimal performance for 5-min segments, achieving an average accuracy of 90%. Interestingly, IMF_FRP features achieved comparable accuracy even for 1-min segments. As segment duration decreased, standard HRV feature accuracy declined while IMF_FRP accuracy stayed intact, eventually matching 5-min segment accuracy levels. The study underscores the surging demand for shorter data segment HRV analysis, driven by advancements in wearable smart watches technology and mobile applications for monitoring health and managing stress. [ABSTRACT FROM AUTHOR]

Facial recognition technology is transformative in security and human-machine interaction, reshaping societal interactions. Robust descriptors, essential for high precision in machine learning tasks like recognition and recall, are integral to this transformation. This paper presents a hybrid model enhancing local binary pattern descriptors for facial representation. By integrating rotation-invariant local binary pattern with uniform rotation-invariant grey-level co-occurrence, employing linear discriminant analysis for feature space optimization, and utilizing an artificial neural network for classification, the model achieves exceptional accuracy rates of 100% for Olivetti Research Laboratory, 99.98% for Maastricht University Computer Vision Test, and 99.17% for Extended Yale B, surpassing traditional methods significantly. [ABSTRACT FROM AUTHOR]

Although linear discriminant analysis (LDA)-based subspace learning has been widely applied to hyperspectral image (HSI) classification, the existing LDA-based subspace learning methods exhibit several limitations: (1) They are often sensitive to noise and demonstrate weak robustness; (2) these methods ignore the local information inherent in data; and (3) the number of extracted features is restricted by the number of classes. To address these drawbacks, this paper proposes a novel joint sparse local linear discriminant analysis (JSLLDA) method by integrating embedding regression and locality-preserving regularization into the LDA model for feature dimensionality reduction of HSIs. In JSLLDA, a row-sparse projection matrix can be learned, to uncover the joint sparse structure information of data by imposing a L 2 , 1 -norm constraint. The L 2 , 1 -norm is also employed to measure the embedding regression reconstruction error, thereby mitigating the effects of noise and occlusions. A locality preservation term is incorporated to fully leverage the local geometric structural information of the data, enhancing the discriminability of the learned projection. Furthermore, an orthogonal matrix is introduced to alleviate the limitation on the number of acquired features. Finally, extensive experiments conducted on three hyperspectral image (HSI) datasets demonstrated that the performance of JSLLDA surpassed that of some related state-of-the-art dimensionality reduction methods. [ABSTRACT FROM AUTHOR]

In the evolving healthcare landscape, recommender systems have gained significant importance due to their role in predicting and anticipating a wide range of health-related data for both patients and healthcare professionals. These systems are crucial for delivering precise information while adhering to high standards of quality, reliability, and authentication. Objectives: The primary objective of this research is to address the challenge of class imbalance in healthcare recommendation systems. This is achieved by improving the prediction and diagnostic capabilities of these systems through a novel approach that integrates linear discriminant wolf (LDW) with convolutional neural networks (CNNs), forming the LDW-CNN model. Methods: The LDW-CNN model incorporates the grey wolf optimizer with linear discriminant analysis to enhance prediction accuracy. The model's performance is evaluated using multi-disease datasets, covering heart, liver, and kidney diseases. Established error metrics are used to compare the effectiveness of the LDW-CNN model against conventional methods, such as CNNs and multi-level support vector machines (MSVMs). Results: The proposed LDW-CNN system demonstrates remarkable accuracy, achieving a rate of 98.1%, which surpasses existing deep learning approaches. In addition, the model improves specificity to 99.18% and sensitivity to 99.008%, outperforming traditional CNN and MSVM techniques in terms of predictive performance. Conclusions: The LDW-CNN model emerges as a robust solution for multidisciplinary disease prediction and recommendation, offering superior performance in healthcare recommender systems. Its high accuracy, alongside its improved specificity and sensitivity, positions it as a valuable tool for enhancing prediction and diagnosis across multiple disease domains. [ABSTRACT FROM AUTHOR]

Featured Application: The developed tool can be applied to motion recognition in individuals to accurately identify the exercises they are performing. This application is particularly useful in fields such as sports science, physical therapy, fitness tracking, and Virtual Reality, where understanding and monitoring exercise routines and movement are crucial. Decision trees are a widely used machine learning technique due to their ease of interpretation and construction. This method allows domain experts to learn from raw data, but they cannot include their prior knowledge in the analysis due to its automatic nature, which implies minimal human intervention in its computation. Conversely, interactive visualization methods have proven to be effective in gaining insights from data, as they incorporate the researcher's criteria into the analysis process. In an effort to combine both methodologies, we have developed a tool to manually build decision trees according to subsequent visualizations of data mapping after applying linear discriminant analysis in combination with Star Coordinates in order to analyze the importance of each feature in the separation. The nodes' information contains data about the features that can be used to split and their cut-off values, in order to select them in a guided manner. In this way, it is possible to produce simpler and more expertly driven decision trees than those obtained by automatic methods. The resulting decision trees reduces the tree size compared to those generated by automatic machine learning algorithms, obtaining a similar accuracy and therefore improving their understanding. The tool developed and presented here to manually create decision trees in a guided manner based on the subsequent visualizations of the data mapping facilitates the use of this method in real-world applications. The usefulness of this tool is demonstrated through a case study with a complex dataset used for motion recognition, where domain experts built their own decision trees by applying their prior knowledge and the visualizations provided by the tool in node construction. The resulting trees are more comprehensible and explainable, offering valuable insights into the data and confirming the relevance of upper body features and hand movements for motion recognition. [ABSTRACT FROM AUTHOR]

Copyright of Shipin Kexue/ Food Science is the property of Food Science Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Accurate detection of stale beef on the market is important for protecting the legitimate rights and interests of consumers. To this end, we combined electronic nose measurements with machine learning technology to classify beef samples. We used an electronic nose to collect information about the odor characteristics of different beef samples and used linear discriminant analysis to reduce data dimensionality. We then classified samples using the following algorithms: extreme gradient boosting, logistic regression, K‐nearest neighbor, random forest, support vector machine, and neural networks for pattern recognition. We assessed model performance using a 10‐fold cross‐validation technique. All these methods reached an accuracy of 95% or above, with F1 scores and AUC values above 0.96. The support vector machine algorithm outperformed all other models, achieving perfect recognition with 100% accuracy and F1/AUC scores of 1.0. Our study demonstrates that electronic nose data combined with support vector machine can be used to successfully discriminate between stale and fresh beef, paving the way for novel research directions in the detection of stale beef. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Food Safety & Quality is the property of Journal of Food Safety & Quality Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Soil spectroscopy offers a method for quantitatively analysing soil chromophores and employing a screening approach. However, utilizing general models with this data at once doesn’t consistently yield accurate results in determining soil parameters. Ambiguities in response within certain domains of the reflectance spectrum are observed. To address this issue, a method is proposed where quantities associated with spectral response are outputted within well-defined domains corresponding to specific parameters. This article introduces a novel approach centred on identifying features from reflectance spectra to assess soil properties and mitigate issues arising from a heterogeneous soil dataset. The focus is on swiftly automating the classification of soil spectral data by extracting features directly related to the varied composition of chromophores in the samples. These features are categorized for each chromophore parameter and analysed using linear discriminant analysis (LDA). The study employs the ‘309 soil samples’ from the GEO-CRADLE open spectral soil library in Greece. Specifically, relevant data from six spectral responses are extracted to separate and quantify certain soil parameters such as clay content, NO3− levels, and SOM (Soil Organic Matter), despite their interrelations. We developed a method to visually represent the determination outcomes quantitatively by establishing three threshold parameter values: 1.6% for SOM, 19% for clay, and 13.6 ppm for NO3−. These determinations will span the explored ranges of ’0.3 to 4.18%’ for SOM, “3 to 48%” for clay, and ’0 to 661.2 ppm’ for NO3−. Utilizing supervised LDA proves notable classification rates of 93.3% for the SOM task, 90% for the Clay task, and 86.6% for NO3− with the specific cut-off value parameter. The final result is presented as a map showing the positions of various prediction samples, highlighting the percentages of clay and organic matter relative to 19% and 1.6%, respectively, as well as the nitrate concentration in these samples relative to 13.6 ppm. [ABSTRACT FROM AUTHOR]

In this paper we address two problems associated with data‐limited dynamic spacecraft exploration: data‐prioritization for transmission, and data‐reduction for interpretation, in the context of ESA ExoMars rover multispectral imaging. We present and explore a strategy for selecting and combining subsets of spectral channels captured from the ExoMars Panoramic Camera, and attempt to seek hematite against a background of phyllosilicates and basalts as a test case scenario, anticipated from orbital studies of the rover landing site. We compute all available dimension reductions on the material reflectance spectra afforded by 4 spectral parameter types, and consider all possible paired combinations of these. We then find the optimal linear combination of each pair whilst evaluating the resultant target‐vs.‐background separation in terms of the Fisher Ratio and classification accuracy, using Linear Discriminant Analysis. We find ∼50,000 spectral parameter combinations with a classification accuracy >95% that use 6‐or‐less filters, and that the highest accuracy score is 99.6% using 6 filters, but that an accuracy of >99% can still be achieved with 2 filters. We find that when the more computationally efficient Fisher Ratio is used to rank the combinations, the highest accuracy is 99.1% using 4 filters, and 95.1% when limited to 2 filters. These findings are applicable to the task of time‐constrained planning of multispectral observations, and to the evaluation and cross‐comparison of multispectral imaging systems at specific material discrimination tasks. Plain Language Summary: Specially designed cameras used by Mars rovers can see not just with the red, green and blue colors of trichromatic vision, but through a dozen or so distinct color channels, some of which extend into the near‐infrared. This super‐human color vision allows for the distinction of a greater diversity of materials, such as types of rocks and soils, than 3‐color vision. This extra color information requires extra data, but there is a limit to the data than can be transmitted from Mars back to Earth each day. If only some of these colors can be transmitted, then which should be chosen? And once transmitted, how should these channels be combined and contrast‐stretched to best convey the content of the scene? That is the problem we address in this paper. We have used mathematical methods from linear algebra to efficiently trial >200,000 possible combinations and contrast stretches of the 12 available color channels of the ExoMars rover Panoramic Camera to find the best combination that uses the smallest number of channels, for finding the water‐related mineral hematite at the landing site of the ExoMars rover. We've found that only 2–6 channels are needed for an accuracy of >99%. Key Points: Method for ranking the ability of multispectral channel combinations to separate target from background material reflectance spectraUsed to seek minimal number of spectral channels needed by the ExoMars Rover PanCam multispectral imager to find hematite at Oxia PlanumFind ∼50,000 combinations with >95% accuracy using six‐or‐less channels, and find that >99% accuracy can be achieved with 2 channels [ABSTRACT FROM AUTHOR]

These days, mobile phones are the most commonly purchased goods. Thousands of new models with improved features, designs, and specifications are released yearly. An autonomous mobile price prediction system is required to assist customers in determining whether or not they can afford these devices. Many machine learning models exhibit varying performance degrees based on their architecture and learning properties. Ten widely used classifiers were assessed in this study: Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Decision Tree (DT), Naïve Bayes (NB), Linear Discriminant Analysis (LDA), AdaBoost, and Light Gradient Boosting (LGB). The F1-score, recall, accuracy, and precision of these models were evaluated. According to the findings, the results indicated that LR, with its use of the Elastic Net parameter, outperformed the others with 96% accuracy, 97% precision, 94% recall, and 96% F1-score. Other models like XGBoost, LGB, and SVM also showed strong performance, whereas KNN had the poorest performance. The study highlights the importance of selecting the appropriate model for accurate mobile price prediction. Among all the machine learning used in this paper, the LR classifier outperforms the other state-of-the-art models because of the elastic Net parameter used for mobile phone price prediction. [ABSTRACT FROM AUTHOR]

Background/Objectives: Head and neck cancer (HNC), predominantly squamous cell carcinoma (SCC), presents a significant global health burden. Conventional diagnostic approaches often face challenges in terms of achieving early detection and accurate diagnosis. This review examines recent advancements in hyperspectral imaging (HSI), integrated with computer-aided diagnostic (CAD) techniques, to enhance HNC detection and diagnosis. Methods: A systematic review of seven rigorously selected studies was performed. We focused on CAD algorithms, such as convolutional neural networks (CNNs), support vector machines (SVMs), and linear discriminant analysis (LDA). These are applicable to the hyperspectral imaging of HNC tissues. Results: The meta-analysis findings indicate that LDA surpasses other algorithms, achieving an accuracy of 92%, sensitivity of 91%, and specificity of 93%. CNNs exhibit moderate performance, with an accuracy of 82%, sensitivity of 77%, and specificity of 86%. SVMs demonstrate the lowest performance, with an accuracy of 76% and sensitivity of 48%, but maintain a high specificity level at 89%. Additionally, in vivo studies demonstrate superior performance when compared to ex vivo studies, reporting higher accuracy (81%), sensitivity (83%), and specificity (79%). Conclusion: Despite these promising findings, challenges persist, such as HSI's sensitivity to external conditions, the need for high-resolution and high-speed imaging, and the lack of comprehensive spectral databases. Future research should emphasize dimensionality reduction techniques, the integration of multiple machine learning models, and the development of extensive spectral libraries to enhance HSI's clinical utility in HNC diagnostics. This review underscores the transformative potential of HSI and CAD techniques in revolutionizing HNC diagnostics, facilitating more accurate and earlier detection, and improving patient outcomes. [ABSTRACT FROM AUTHOR]

This article provides an overview of modern machine learning methods in the context of their active use in credit scoring, with particular attention to the following algorithms: light gradient boosting machine (LGBM) classifier, logistic regression (LR), linear discriminant analysis (LDA), decision tree (DT) classifier, gradient boosting classifier and extreme gradient boosting (XGB) classifier. Each of the methods mentioned is subject to careful analysis to evaluate their applicability and effectiveness in predicting credit risk. The article examines the advantages and limitations of each method, identifying their impact on the accuracy and reliability of borrower creditworthiness assessments. Current trends in machine learning and credit scoring are also covered, warning of challenges and discussing prospects. The analysis highlights the significant contributions of methods such as LGBM classifier, LR, LDA, DT classifier, gradient boosting classifier and XGB classifier to the development of modern credit scoring practices, highlighting their potential for improving the accuracy and reliability of borrower creditworthiness forecasts in the financial services industry. Additionally, the article discusses the importance of careful selection of machine learning models and the need to continually update methodology in light of the rapidly changing nature of the financial market. [ABSTRACT FROM AUTHOR]

Abstract This research paper introduces an innovative approach to classify heart rate variability (HRV) time series into paced and spontaneous breathing patterns to reflect changes in the autonomic nervous system. This type of classification is beneficial in wearable devices for stress/relaxation level detection and in deciding therapeutic interventions. The “Multi-Domain Approach” methodology integrates three different techniques: standard HRV features, fuzzy recurrence plot (FRP)-based FRP_GLCM, and empirical mode decomposition-based IMF_FRP_GLCM. The study concentrates on analyzing HRV time series within shorter data segments, aligning with the requirements of contemporary wearable health devices and biofeedback systems. HRV data collected during spontaneous and slow-paced breathing were analyzed across data segments of 5, 4, 3, 2, and 1 min, incorporating feature selection and reduction methods. Results demonstrated that standard HRV features yielded optimal performance for 5-min segments, achieving an average accuracy of 90%. Interestingly, IMF_FRP features achieved comparable accuracy even for 1-min segments. As segment duration decreased, standard HRV feature accuracy declined while IMF_FRP accuracy stayed intact, eventually matching 5-min segment accuracy levels. The study underscores the surging demand for shorter data segment HRV analysis, driven by advancements in wearable smart watches technology and mobile applications for monitoring health and managing stress.

Lipstick stains discovered on objects (such as drinking cups) may be utilized as evidence and/or lead in forensic investigations by chemical analysis. Hence, this study analyzed two brands of lipstick stains' organic composition on paper cups over 0, 24, and 48 h of exposure using attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR), coupled with pattern recognition techniques of principal component analysis (PCA) and linear discriminant analysis (LDA). Spectral analysis revealed several functional group differences between the two brands of lipstick, viz. Wardah (W) and Silkygirl (S) stains, particularly in the fingerprint region of 1800–650 cm−1. Application of PCA (variance of ≥ 70%) and LDA successfully provided the suitable means for categorically classifying (a) the different brands of lipstick, (b) the male and female volunteers, as well as (c) the different periods of exposure. The fact that the cross-validation correct classification rates for all the LDA models exceeded 90.0%, their suitability for forensic provenance application of lip stains appears statistically supported. Hence, such an approach to combine the non-destructive ATR-FTIR analysis with PCA and LDA is not only easy and cost-effective but also forensically relevant for enabling the data mining process for forensic intelligence.

This study investigated the combined application of visible/near-infrared (Vis-NIR) spectroscopy with linear discriminant analysis (LDA) and machine learning (ML) for the classification of apples with different degrees of damage. The Vis-NIR spectral data of apples with different degrees of damage were collected, and the effect of different spectral preprocessing methods on the support vector machine (SVM) classification model was analyzed. LDA was used to reduce the dimensionality of the preprocessed spectral data, and five machine learning models including SVM, random forest (RF), K-nearest neighbor (KNN), decision tree (DT) and extreme gradient boosting (XGBoost) were constructed and compared for the classification of apple damage. The results showed that the SVM model based on preprocessed spectra with Savitzky-Golay (SG) smoothing had the best classification performance, with an accuracy of 87.3%. After dimensionality reduction using LDA, the classification accuracy of all the models was significantly improved, with the highest increase of 16% being observed in the DT model. The KNN model showed the best classification performance, with an accuracy of 96.0% and a precision of 96.4%. This study provides a basis for efficient and accurate assessment of the degree of mechanical damage in apples.

These days, mobile phones are the most commonly purchased goods. Thousands of new models with improved features, designs, and specifications are released yearly. An autonomous mobile price prediction system is required to assist customers in determining whether or not they can afford these devices. Many machine learning models exhibit varying performance degrees based on their architecture and learning properties. Ten widely used classifiers were assessed in this study: Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), Decision Tree (DT), Naïve Bayes (NB), Linear Discriminant Analysis (LDA), AdaBoost, and Light Gradient Boosting (LGB). The F1-score, recall, accuracy, and precision of these models were evaluated. According to the findings, the results indicated that LR, with its use of the Elastic Net parameter, outperformed the others with 96% accuracy, 97% precision, 94% recall, and 96% F1-score. Other models like XGBoost, LGB, and SVM also showed strong performance, whereas KNN had the poorest performance. The study highlights the importance of selecting the appropriate model for accurate mobile price prediction. Among all the machine learning used in this paper, the LR classifier outperforms the other state-of-the-art models because of the elastic Net parameter used for mobile phone price prediction.

Abstract The challenges associated with liquid biopsy of colorectal cancer (CRC) are closely linked to the substantial variations observed in gene expression profiles among patients. This variability complicates the selection of an ideal biomarker for accurate diagnosis. In this report, we propose that employing a combination of miRNAs offers a better change for enhancing the accuracy of CRC diagnosis compared to solely relying on single miRNAs. As an illustrative example, we measured 9 miRNAs from 45 patient samples (comprising 31 CRC cases and 14 healthy controls) via RT-qPCR. We then utilized two methods: (1) LASSO regression for marker ranking and (2) linear discriminant analysis (LDA) to identify the optimal weighted combination of multiple markers. Our data indicates that combination of triple markers, selected based on their ranking, exhibited the highest diagnostic performance, including a sensitivity of 93.6% (95% confidence interval, CI 79.3–98.9%), specificity of 100% (CI 78.5–100.0%), positive predictive value (PPV) of 100%, negative predictive value (NPV) of 87.5%, and an overall accuracy of 95.6%. In contrast, the diagnostic performance of each individual miRNA used in the triple marker combination ranged from 53.3 to 80.0% in accuracy. While we acknowledge the need for further extensive studies involving larger patient cohorts and the consideration of additional miRNA candidates, our research undeniably highlights the potential of combining multiple markers as a robust methodology for identifying biomarkers among heterogeneous patient profiles.

Training deep Convolutional Neural Networks consists of multiple forward and backward passes over several epochs until the loss converges, making it time-consuming. Various factors affect the training times of networks, weight initialization being one of them. The idea of a proper weight initialization technique is to set the initial weights such that the network converges faster by extracting meaningful features from the data. This paper proposes a data-driven weight initialization to accelerate the training process. This technique is based on obtaining initial weights by appropriately analysing the training data. The proposed weight initialization strategy initializes filters from a pre-defined filter bank that is created before training, and it contains standard edge and texture feature extracting filters and data-driven filters obtained using principal component analysis, linear discriminant analysis and partial least squares. The proposed technique has been validated on AlexNet, VGG-16 and ResNet-50 for Intel Image Classification, CIFAR10 and CIFAR100 datasets. The results show that the proposed technique gives better validation results than other state-of-the-art techniques in fewer epochs and works well with state-of-the-art activation functions. [ABSTRACT FROM AUTHOR]

This study explores the field of sign language recognition through machine learning, focusing on the development and comparative evaluation of various algorithms designed to interpret sign language. With the prevalence of hearing impairment affecting millions globally, efficient sign language recognition systems are increasingly critical for enhancing communication for the deaf and hard-of-hearing community. We review several studies, showcasing algorithms with accuracies ranging from 63.5 to 99.6%. Building on these works, we introduce a novel algorithm that has been rigorously tested and has demonstrated a perfect accuracy of 99.7%. Our proposed algorithm utilizes a sophisticated convolutional neural network architecture that outperforms existing models. This work details the methodology of the proposed system, which includes preprocessing, feature extraction, and a multi-layered CNN approach. The remarkable performance of our algorithm sets a new benchmark in the field and suggests significant potential for real-world application in assistive technologies. We conclude by discussing the impact of these findings and propose directions for future research to further improve the accessibility and effectiveness of sign language recognition systems.

Abstract This work proposes a functional data analysis approach for morphometrics in classifying three shrew species (S. murinus, C. monticola, and C. malayana) from Peninsular Malaysia. Functional data geometric morphometrics (FDGM) for 2D landmark data is introduced and its performance is compared with classical geometric morphometrics (GM). The FDGM approach converts 2D landmark data into continuous curves, which are then represented as linear combinations of basis functions. The landmark data was obtained from 89 crania of shrew specimens based on three craniodental views (dorsal, jaw, and lateral). Principal component analysis and linear discriminant analysis were applied to both GM and FDGM methods to classify the three shrew species. This study also compared four machine learning approaches (naïve Bayes, support vector machine, random forest, and generalised linear model) using predicted PC scores obtained from both methods (a combination of all three craniodental views and individual views). The analyses favoured FDGM and the dorsal view was the best view for distinguishing the three species.

Ensuring software reliability before public release is crucial, as most issues stem from human errors during development. Addressing these early through integrated testing resources can mitigate potential problems. This paper proposes an enhanced software fault prediction model using a meta-heuristic optimization technique. Our approach utilizes NASA’s datasets, incorporating data cleaning, feature dimensional reduction, and software fault prediction (SFP). The Extreme Learning Machine (ELM) parameters, such as weights and biases, were optimally determined using the improved particle swarm optimization (IMPSO) method. Model validation employed various metrics, including accuracy, sensitivity, specificity, F1 score, and MCC, through a 10 × 5 cross-validation. The model, named PCALDA+IMPSO)ELM, was tested on NASA datasets (CM1, KC2, KC3, MC1, and PC1), achieving prediction accuracies of 0.9696, 0.9836, 0.9482, 0.9799, and 0.9644, respectively. The findings demonstrate significant promise for the meta-heuristic optimization technique in software fault prediction with reduced feature sets, offering an effective method to enhance software reliability and anticipate fault-prone modules. [ABSTRACT FROM AUTHOR]

Copyright of Rare Metals is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Brain tumors pose significant threats within neurological disorders, demanding accurate classification for effective diagnosis and treatment. This study explores brain tumor classification employing Classical Local Binary Patterns (CLBP) and Convolutional Neural Networks (CNN), alongside texture feature extraction from MRI images using classical LBP and HOG (Histogram of Oriented Gradients). These methods adeptly capture both local and global texture patterns crucial for tumor identification. Our proposed framework encompasses three pivotal steps: image pre-processing, feature extraction via CLBP, and classification utilizing CNN. Evaluation on a publicly available brain tumor dataset showcased an impressive 95.6% accuracy in tumor classification, affirming the efficacy of the CLBP+CNN approach. This method bears promising implications for enhancing clinical diagnosis and treatment planning. Furthermore, we propose future extensions including CLBPs such as DLBP and LBP. DLBP introduces a parameter, 'D', dictating pixel distance, while LBP varies pixel values across specified ranges. Additionally, tumor classification was explored employing ANN, AlDE, and LDA classification methods, with future prospects of incorporating DLBP, LBP, and CLBP extractions from MRI images within the dataset [ABSTRACT FROM AUTHOR]

Linear discriminant analysis (LDA) is widely used for dimensionality reduction under supervised learning settings. Traditional LDA objective aims to minimize the ratio of the squared Euclidean distances that may not perform optimally on noisy datasets. Multiple robust LDA objectives have been proposed to address this problem, but their implementations have two major limitations. One is that their mean calculations use the squared \(\ell_{2}\) -norm distance to center the data, which is not valid when the objective depends on other distance functions. The second problem is that there is no generalized optimization algorithm to solve different robust LDA objectives. In addition, most existing algorithms can only guarantee that the solution is locally optimal rather than globally optimal. In this article, we review multiple robust loss functions and propose a new and generalized robust objective for LDA. Besides, to remove the mean value within data better, our objective uses an optimal way to center the data through learning. As one important algorithmic contribution, we derive an efficient iterative algorithm to optimize the resulting non-smooth and non-convex objective function. We theoretically prove that our solution algorithm guarantees that both the objective and the solution sequences converge to globally optimal solutions at a sub-linear convergence rate. The results of comprehensive experimental evaluations demonstrate the effectiveness of our new method, achieving significant improvements compared to the other competing methods. [ABSTRACT FROM AUTHOR]

Aiming to improve the robustness and adaptiveness of the recently investigated capped norm linear discriminant analysis (CLDA), this paper proposes a regularized linear discriminant analysis based on the generalized capped l 2 , q -norm (GCLDA). Compared to CLDA, there are two improvements in GCLDA. Firstly, GCLDA uses the capped l 2 , q -norm rather than the capped l 2 , 1 -norm to measure the within-class and between-class distances for arbitrary q > 0 . By selecting an appropriate q, GCLDA is adaptive to different data, and also removes extreme outliers and suppresses the effect of noise more effectively. Secondly, by taking into account a regularization term, GCLDA not only improves its generalization ability but also avoids singularity. GCLDA is solved through a series of generalized eigenvalue problems. Experiments on an artificial dataset, some real world datasets and a high-dimensional dataset demonstrate the effectiveness of GCLDA. [ABSTRACT FROM AUTHOR]

Summary: This study investigated the variation in mineral elements in bracatinga (Mimosa scabrella Bentham) honeydew honeys produced biannually over three consecutive harvest years. The aim was to assess the relationship between the year of production and the concentration of fourteen rare earth elements and six trace elements analysed using ICP‐MS. The concentrations of the twenty elements varied (P < 0.001) among all the honey samples. The dominant rare earth elements were Eu, followed by Nd and La, while Pd and Au were the major trace elements. Cluster analysis (CA) and linear discriminant analysis (LDA) were performed using all the elements identified. The LDA categorised (83.3% accuracy) the bracatinga honeydew honey samples according to their harvest year. Discriminant functions, influenced by Ce, Tb, and Th, show distinct elemental patterns in bracatinga honeydew honeys, varying with harvest year. This reflects fluctuations in yearly climatic conditions and is important to guarantee accurate characterisation. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a major global health concern that affects millions of people globally. This study investigates the technical challenges in AD analysis and provides a thorough analysis of AD, emphasizing the disease's worldwide effects as well as the predicted increase. It explores the technological difficulties associated with AD analysis, concentrating on the shift in automated clinical diagnosis using MRI data from conventional machine learning to deep learning techniques. This study advances our knowledge of the effects of AD and provides new developments in deep learning for precise diagnosis, providing insightful information for both clinical and future research. The research introduces an innovative deep learning model, leveraging YOLOv5 and variants of YOLOv8, to classify AD images into four (NC, EMCI, LMCI, AD) categories. This study evaluates the performance of YOLOv5 which achieved high accuracy (97%) in multi‐class classification (classes 0 to 3) with precision, recall, and F1‐score reported for each class. YOLOv8 (Small) and YOLOv8 (Medium) models are also assessed for Alzheimer's disease diagnosis, demonstrating accuracy of 97% and 98%, respectively. Precision, recall, and F1‐score metrics provide detailed insights into the models' effectiveness across different classes. Comparative analysis against a transfer learning model reveals YOLOv5, YOLOv8 (Small), and YOLOv8 (Medium) consistently outperforming across six binary classifications related to cognitive impairment. These models show improved sensitivity and accuracy compared to baseline architectures from [32]. In AD/NC classification, YOLOv8 (Medium) achieves 98.43% accuracy and 97.45% sensitivity, for EMCI/LMCI classification, YOLOv8 (Medium) also excels with 92.12% accuracy and 90.12% sensitivity. The results highlight the effectiveness of YOLOv5 and YOLOv8 variants in neuroimaging tasks, showcasing their potential in clinical applications for cognitive impairment classification. The proposed models showcase superior performance, achieving high accuracy, sensitivity, and F1‐scores, surpassing baseline architectures and previous methods. Comparative analyses highlight the robustness and effectiveness of the proposed models in AD classification tasks, providing valuable insights for future research and clinical applications. [ABSTRACT FROM AUTHOR]

Emotion based music player is an interdisciplinary study of computer vision and psychology. As music enhances the positive vibes it plays a significant role in soothing people's emotion. Emotions can be predicted through facial expression analysis using vision-based methods. However, challenges like environment and expression complexity have become hindrance to attain a good recognition rate. Therefore, we put forward a deep dual domain joint feature framework based on linear discriminant analysis for facial emotion recognition. First, we detect the human face and learn the emotion pattern using the popular complementary deep domain networks called EfficientNet and ResNet50. The learned deep dual domain space is projected onto linear discriminant space to achieve a joint discriminant feature space. The recognition rate of the proposed joint discriminant feature framework is analyzed using support vector machine. To prove the efficacy of the proposed framework, we validated it on two Benchmarks namely FER2013 and CK48+ datasets. The proposed framework achieved a good recognition rate of 99% and 98.6% on FER2013 and CK48+ respectively. Experimental analysis on our EmDe dataset showed an accuracy of 99% and proves that the deep dual domain joint discriminant framework as a promising pipeline for emotion-based music player system. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Food Safety & Quality is the property of Journal of Food Safety & Quality Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract.In this work, we consider the discriminant analysis under the weak sparsity where many entries of the parameters are nearly zero. We develop a unified LASSO-typed framework to estimate the parameters for sparse discriminant analysis and derive the general non asymptotic error bound under the weak sparsity condition. As applications, we revisit the sparse linear discriminant analysis and sparse quadratic discriminant analysis. We establish the consistency of the estimators and also the misclassification error rate. These results extend the sparse discriminant analysis methods to weak sparsity setting and refine the existing theoretical results. [ABSTRACT FROM AUTHOR]

Key message: Hyperspectral features enable accurate classification of soybean seeds using linear discriminant analysis and GWAS for novel seed trait genes. Evaluating crop seed traits such as size, shape, and color is crucial for assessing seed quality and improving agricultural productivity. The introduction of the SUnSet toolbox, which employs hyperspectral sensor-derived image analysis, addresses this necessity. In a validation test involving 420 seed accessions from the Korean Soybean Core Collections, the pixel purity index algorithm identified seed- specific hyperspectral endmembers to facilitate segmentation. Various metrics extracted from ventral and lateral side images facilitated the categorization of seeds into three size groups and four shape groups. Additionally, quantitative RGB triplets representing seven seed coat colors, averaged reflectance spectra, and pigment indices were acquired. Machine learning models, trained on a dataset comprising 420 accession seeds and 199 predictors encompassing seed size, shape, and reflectance spectra, achieved accuracy rates of 95.8% for linear discriminant analysis model. Furthermore, a genome-wide association study utilizing hyperspectral features uncovered associations between seed traits and genes governing seed pigmentation and shapes. This comprehensive approach underscores the effectiveness of SUnSet in advancing precision agriculture through meticulous seed trait analysis. [ABSTRACT FROM AUTHOR]

Redox-sensitive cysteine (RSC) thiol plays an important role in many biological processes such as photosynthesis, cellular metabolism, and transcription. Therefore, it is necessary to identify redox-sensitive cysteine accurately. However, traditional redox-sensitive cysteine identification is very expensive and time-consuming. At present, there is an urgent need for a mathematical calculation method to identify sequence information and redox-sensitive cysteines quickly and accurately. Here, we developed an effective predictor called iRSC-PseAAC, which used the dimension reduction algorithm LDA combined with the support vector machine to predict redox-sensitive cysteine sites. In the cross-validation results, the specificity (Sp), sensitivity (Sn), accuracy (Acc) and Matthews correlation coefficient (MCC) were 0.841, 0.868, 0.859 and 0.692 respectively. In the independent dataset results, the Sp, Sn, Acc and MCC were 0.906, 0.882, 0.890 and 0.767 respectively. compared with existing prediction methods, iRSC-PseAAC had obvious improvement effect. The method proposed for this study can also be used for many problems in computational proteomics. [ABSTRACT FROM AUTHOR]

Flax hemp and ramie fibers have been important raw materials for textile products since ancient times. Due to the difference in use and output there has been a significant gap in the market price of these three different species of bast fibers in recent years and the illegal phenomenon of low-price bast fibers posing as high-price bast fibers often occurs. The composition physical and chemical properties of flax hemp and ramie fibers are very close. At present the standard method for identifying bast fibers in China is FZ/ T 01057. 3 2007 The Method for Identification of Textile Fibers Part 3 Microscopy. However the microscopic morphology of flax hemp and ramie is highly similar and the accuracy of identifying bast fibers according to this method is insufficient. The identification of these three species of easily confused bast fibers has always been a difficult task in the field of textile inspection and testing so it is necessary to improve the identification technology of bast fibers to maintain market stability and strengthen quality supervision. Although the predecessors have found that there are relatively slight infrared spectral differences between different kinds of bast fibers no rapid and accurate classification method for bast fibers based on the differences in infrared spectral data has been reported. In our study the principal component analysis PCA and linear discriminant analysis LDA on attenuated total reflection Flourier transformed infrared spectroscopy ATR-FTIR of different kinds of bast fibers were proposed to establish a discriminant model to identify these three kinds of easily confused bast fibers. In this study 60 groups of flax hemp and ramie fibers were selected as sample sets for degumming cleaning treatment and their ATR-FTIR spectra were collected. After spectral normalization infrared spectra in the range of 800 - 2 000 cm- 1 were compressed to scores of sets of principal components PCs by PCA. The PCA results show that with the increase of the number of PCs the PC scores gradually display a clustering trend according to the species of bast fibers and the cumulative variance contribution rate of the first 12 PCs to the normalized spectral data reaches 99. 5% . With the first 12 PC scores as the independent variable and the species of fiber samples as the dependent variable a discriminant model of the bast fibers including typical discriminant functions and classification functions was established by LDA. The discriminant model verification results show that the typical discriminant functions can make the first 12 PC scores achieve a excellent cluster according to the species of fiber samples and the classification functions can reach a classification accuracy of 100% for all fiber samples in the training set and the test set. In addition the classification accuracy of the PCA-LDA model can still reach 99. 6% by leave-one-out cross-validation. These results prove that the differences between ATR-FTIR spectra of flax hemp and ramie can be discerned and utilized for the rapid and accurate identification of these easily confused textile bast fibers by PCA-LDA statistic analysis. In the follow-up study we will introduce more laboratories and samples to participate in the construction and verification of the PCA-LDA model so as to further improve the applicability of the bast fiber classification and identification model. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to assess the diagnostic power of brain asymmetry indices and neuropsychological tests for differentiating mesial temporal lobe epilepsy (MTLE) and schizophrenia (SCZ). Methods: We studied a total of 39 women including 13 MTLE, 13 SCZ, and 13 healthy individuals (HC). A neuropsychological test battery (NPT) was administered and scored by an experienced neuropsychologist, and NeuroQuant (CorTechs Labs Inc., San Diego, California) software was used to calculate brain asymmetry indices (ASI) for 71 different anatomical regions of all participants based on their 3D T1 MR imaging scans. Results: Asymmetry indices measured from 10 regions showed statistically significant differences between the three groups. In this study, a multi-class linear discriminant analysis (LDA) model was built based on a total of fifteen variables composed of the most five significantly informative NPT scores and ten significant asymmetry indices, and the model achieved an accuracy of 87.2%. In pairwise classification, the accuracy for distinguishing MTLE from either SCZ or HC was 94.8%, while the accuracy for distinguishing SCZ from either MTLE or HC was 92.3%. Conclusion: The ability to differentiate MTLE from SCZ using neuroradiological and neuropsychological biomarkers, even within a limited patient cohort, could make a substantial contribution to research in larger patient groups using different machine learning techniques. [ABSTRACT FROM AUTHOR]

Copper-cobalt bimetallic nitrogen-doped carbon-based nanoenzymatic materials (CuCo@NC) were synthesized using a one-step pyrolysis process. A three-channel colorimetric sensor array was constructed for the detection of seven antioxidants, including cysteine (Cys), uric acid (UA), tea polyphenols (TP), lysine (Lys), ascorbic acid (AA), glutathione (GSH), and dopamine (DA). CuCo@NC with peroxidase activity was used to catalyze the oxidation of TMB by H2O2 at three different ratios of metal sites. The ability of various antioxidants to reduce the oxidation products of TMB (ox TMB) varied, leading to distinct absorbance changes. Linear discriminant analysis (LDA) results showed that the sensor array was capable of detecting seven antioxidants in buffer and serum samples. It could successfully discriminate antioxidants with a minimum concentration of 10 nM. Thus, multifunctional sensor arrays based on CuCo@NC bimetallic nanoenzymes not only offer a promising strategy for identifying various antioxidants but also expand their applications in medical diagnostics and environmental analysis of food. [ABSTRACT FROM AUTHOR]

Enhancing user experience in music listening through facial emotion recognition presents a novel avenue for curating personalized song playlists, obviating the need for traditional search methodologies like keyword inputs or manual playlist exploration. This research pivots towards leveraging advanced pre-trained models, specifically ResNet18 and VGG19, for the accurate detection and analysis of listeners' emotions via facial expressions. By integrating these models with the Facial Emotion Recognition (FER) 2013, and The Extended Cohn-Kanade Dataset (CK+) dataset, our study aims to precisely identify a range of emotions including happiness, anger, sadness, surprise, disgust, and neutrality. In parallel, we employ Support Vector Machine (SVM) and Linear Discriminant Analysis (LDA) techniques to discern musical emotions from Spotify's dataset, encompassing moods such as happiness, sadness, energy, and calmness. A unique set of rules is formulated to amalgamate the insights from facial emotion recognition with musical emotion data, facilitating the automated suggestion of playlists that resonate with the user's current emotional state. Our experiments demonstrate notable results with the ResNet18 model achieved 72.67% accuracy and a 72.30% F1 score on the FER2013 dataset, with significant improvements on the CK+ dataset, reaching 96.97% accuracy and a 96.60% F1 score. The VGG19 model similarly excelled, particularly on the CK+ dataset with 95.96% accuracy and a 96.60% F1 score. On the FER2013 dataset, the VGG19 model achieved an accuracy of 72.64%, and an F1 score of 72.60%. Our music emotion recognition approach, using SVM on Spotify's dataset, achieved an 86.96% accuracy and an 84.54% F1 score. When using the LDA model on the same dataset, the accuracy was 79.00% and an F1 score of 76.72%. To bring this concept to fruition, we have developed an intuitive web application that allows users to capture their facial expressions via their device's camera, subsequently offering song recommendations tailored to their detected mood. [ABSTRACT FROM AUTHOR]

The sparse linear discriminant analysis can be incorporated into the penalized linear regression framework, but most studies have been limited to specific convex penalties, including the least absolute selection and shrinkage operator and its variants. Within this framework, concave penalties can serve as natural counterparts of the convex penalties. Implementing the concave penalized direction vector of discrimination appears to be straightforward, but developing its theoretical properties remains challenging. In this paper, we explore a class of concave penalties that covers the smoothly clipped absolute deviation and minimax concave penalties as examples. We prove that employing concave penalties guarantees an oracle property uniformly within this penalty class, even for high-dimensional samples. Here, the oracle property implies that an ideal direction vector of discrimination can be exactly recovered through concave penalized least squares estimation. Numerical studies confirm that the theoretical results hold with finite samples. [ABSTRACT FROM AUTHOR]

Universum linear discriminant analysis was recently proposed to improve linear discriminant analysis by incorporating Universum information. However, it obtains each of the discriminant directions by just using samples from two classes, while other class samples are considered as Universum. This not only leads to the ignoring of some discriminant information, but also restricts its number of extracted features to at most 0.5 k (k - 1) , where k is the number of classes. To fully explore discriminant information from all classes, this paper studies a novel Universum linear discriminant analysis by considering a unified model that simultaneously uses all classes. Compared to the existing Universum linear discriminant analysis, all Universum information is fully utilized in the proposed model when obtaining each discriminant direction, where the Universum can be self-constructed as well can be given advanced of any types. The constrained "concave-convex" procedure can be used to solve the proposed method, which makes the algorithm convergent to a local minimum. By employing a recursive technique, arbitrary number of discriminant directions can be obtained. Experimental results on real-world benchmark datasets and image datasets illustrate the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

The existing clinical diagnostic methods of pulmonary tuberculosis (TB) usually have some of the following limitations, such as time-consuming, invasive, radioactive, insufficiently sensitive and accurate. This study demonstrates the possibility of using blood plasma autofluorescence spectroscopy and Artificial Neural Network (ANN) algorithm for the rapid and accurate diagnosis of latent and active pulmonary TB from healthy subjects. The fluorescence spectra of blood plasma from 18 healthy volunteers, 12 individuals with latent TB infections and 80 active TB patients are measured and analyzed. By optimizing the ANN structure and activation functions, the ANN three-classification model achieves average classification accuracy of 96.3 %, and the accuracy of healthy persons, latent TB infections and active TB patients are 100 %, 83.3 % and 97.5 %, respectively, which is much better than the results of traditional Principal component analysis (PCA) and Linear discriminant analysis (LDA) method. To the best of our knowledge, this is the first research work of differentiating latent, active pulmonary TB cases from healthy samples with autofluorescence spectroscopy. As a rapid, accurate, safe, label-free, non-invasive and cost-effective technique, it can be developed as a promising diagnostic tool for the screening of pulmonary TB disease in the early stage.

With the advent of the information age, to provide better services and ensure the security management of libraries, intelligent facial recognition technology has gradually become a hot research direction in library management. Meanwhile, to further improve the comprehensive performance of facial recognition, this study attempts to integrate principal component analysis and linear discriminant analysis on the basis of analyzing the framework of recognition technology. Afterwards, it introduced support vector machines for recognition and classification, and proposed a new recognition model. The experimental results show that the recognition accuracy of the proposed model is up to 97 % in the ORL dataset and 94 % in the Yale dataset. The recognition error rate is as low as 0.1 % when the number of training samples is 215 and the number of iterations is 200. The model has the best recognition performance when the image size is 25 × 25 mm and the number of noises is 10. In addition, the model is particularly effective in recognition on single person color or gray images, with the highest P-value of 98.7 %, the highest R-value of 98.8 %, and the highest F1-value of 97.5 %. These results show that the proposed model significantly improves the accuracy and robustness of face recognition, and provides strong technical support for intelligent service innovation in smart libraries.

We hydroponically cultivated wheat and subjected them to mineral deficiency by a week-long exposure to Hoagland's solution minus one essential mineral. Chlorophyll fluorescence and P700+ absorbance were measured under a single 300 ms saturating light pulse (one-shot) or the conventional two pulses (two-shot). A machine learning-based multiclass classification model was developed to assess mineral deficiency diagnosis. The results indicated that partial least squares discriminant analysis (PLS-DA) was ineffective. Conversely, linear discriminant analysis (LDA) without principal components analysis (PCA) successfully identified 13 or 11 (with a class combining the control with Mo- and B-deficiencies; and ten other distinct classes for respective mineral deficiencies) classes of mineral deficiencies, achieving over 81 % diagnostic performance across all five macro metrics of precision, recall, specificity, accuracy, and F1 measure. The one-shot method exhibited only a slight decrease in performance compared to the two-shot method, highlighting the advantage of substantially reduced measurement time. Applying the developed LDA models to outdoor soil-grown leaf data revealed their practicality and limitations. In multiple deficiency scenarios, the model predicted one predominant deficiency or incorrectly diagnosed a different deficiency. This study is the first to quantitatively demonstrate the efficacy of ultra-fast nutritional diagnostic technology in identifying root zone mineral deficiencies using chlorophyll fluorescence and P700+ absorbance measurements. It delineates the requirements for constructing multilabel or quantitative calibration models for field deployment.

Abstract In this paper we address two problems associated with data‐limited dynamic spacecraft exploration: data‐prioritization for transmission, and data‐reduction for interpretation, in the context of ESA ExoMars rover multispectral imaging. We present and explore a strategy for selecting and combining subsets of spectral channels captured from the ExoMars Panoramic Camera, and attempt to seek hematite against a background of phyllosilicates and basalts as a test case scenario, anticipated from orbital studies of the rover landing site. We compute all available dimension reductions on the material reflectance spectra afforded by 4 spectral parameter types, and consider all possible paired combinations of these. We then find the optimal linear combination of each pair whilst evaluating the resultant target‐vs.‐background separation in terms of the Fisher Ratio and classification accuracy, using Linear Discriminant Analysis. We find ∼50,000 spectral parameter combinations with a classification accuracy >95% that use 6‐or‐less filters, and that the highest accuracy score is 99.6% using 6 filters, but that an accuracy of >99% can still be achieved with 2 filters. We find that when the more computationally efficient Fisher Ratio is used to rank the combinations, the highest accuracy is 99.1% using 4 filters, and 95.1% when limited to 2 filters. These findings are applicable to the task of time‐constrained planning of multispectral observations, and to the evaluation and cross‐comparison of multispectral imaging systems at specific material discrimination tasks.