Learning Classifiers Research Articles

Quantum data encoding: a comparative analysis of classical-to-quantum mapping techniques and their impact on machine learning accuracy

This study explores the integration of quantum data embedding techniques into classical machine learning (ML) algorithms; to assess performance enhancements and computational implications across a spectrum of models. We explored various classical-to-quantum mapping methods; ranging from basis encoding and angle encoding to amplitude encoding; for encoding classical data. We conducted an extensive empirical study encompassing popular ML algorithms, including Logistic Regression, K-Nearest Neighbors, Support Vector Machines, and ensemble methods like Random Forest, LightGBM, AdaBoost, and CatBoost. Our findings reveal that quantum data embedding contributes to improved classification accuracy and F1 scores, particularly notable in models that inherently benefit from enhanced feature representation. We observed nuanced effects on running time, with low-complexity models exhibiting moderate increases and more computationally intensive models experiencing discernible changes. Notably, ensemble methods demonstrated a favorable balance between performance gains and computational overhead.This study underscores the potential of quantum data embedding to enhance classical ML models and emphasizes the importance of weighing performance improvements against computational costs. Future research may involve refining quantum encoding processes to optimize computational efficiency and explore scalability for real-world applications. Our work contributes to the growing body of knowledge on the intersection of quantum computing and classical machine learning, offering insights for researchers and practitioners seeking to harness the advantages of quantum-inspired techniques in practical scenarios.

A multivariate analysis of the early dropout using classical machine learning and local interpretable model-agnostic explanations

Student dropout rates can have a significant negative impact on both the development of educational institutions and the personal growth of students. Consequently, many institutions are focused on identifying key factors that contribute to dropout and implementing strategies to mitigate them. This study aims to predict student dropout rates using classical machine learning algorithms while analyzing the key factors influencing these outcomes in higher education. The dataset includes demographic, socioeconomic, and academic information from various sources. Additionally, the study leverages the Local Interpretable Model-Agnostic Explanations (LIME) model to provide insights into the predictions, offering a clearer understanding of the factors driving dropout decisions. This knowledge is crucial for identifying influential factors and, more importantly, enhancing early intervention strategies and policies in educational settings, ultimately reducing dropout rates.

Explainable Multi-Layer Dynamic Ensemble Framework Optimized for Depression Detection and Severity Assessment

Depression is a common mental health issue, especially among older adults, and early detection is critical to effective intervention and treatment. This study proposes an explainable multilayer dynamic ensemble framework optimized for detecting depression and predicting its severity. Using data from real individuals from the National Social Life, Health, and Aging Project (NSHAP), the framework integrates classical machine learning models, static ensemble techniques, and dynamic ensemble selection (DES) approaches to build a two-stage framework. The depression detection process is performed in the first stage, and the depression severity is predicted in the second stage only for depressed patients. Among the models evaluated, the FIRE-KNOP DES algorithm exhibited the highest performance, achieving 88.33% accuracy in detecting depression and 83.68% accuracy in predicting depression severity. A key innovation of this study is the incorporation of explainable AI (XAI) techniques to enhance model interpretability, making the framework more suitable for clinical applications. We explore different global and local XAI features. Model explainability highlights the significance of medically relevant mental and non-mental health features in improving the model’s performance, which enhances the trustworthiness of the domain experts in the model decisions. These findings offer a promising foundation for future applications of dynamic ensemble frameworks in mental health assessments.

Airborne LiDAR Point Cloud Classification Using Ensemble Learning for DEM Generation

Airborne laser scanning (ALS) point clouds have emerged as a predominant data source for the generation of digital elevation models (DEM) in recent years. Traditionally, the generation of DEM using ALS point clouds involves the steps of point cloud classification or ground point filtering to extract ground points and labor-intensive post-processing to correct the misclassified ground points. The current deep learning techniques leverage the ability of geometric recognition for ground point classification. However, the deep learning classifiers are generally trained using 3D point clouds with simple geometric terrains, which decrease the performance of model inferencing. In this study, a point-based deep learning model with boosting ensemble learning and a set of geometric features as the model inputs is proposed. With the ensemble learning strategy, this study integrates specialized ground point classifiers designed for different terrains to boost classification robustness and accuracy. In experiments, ALS point clouds containing various terrains were used to evaluate the feasibility of the proposed method. The results demonstrated that the proposed method can improve the point cloud classification and the quality of generated DEMs. The classification accuracy and F1 score are improved from 80.9% to 92.2%, and 82.2% to 94.2%, respectively, by using the proposed methods. In addition, the DEM generation error, in terms of mean squared error (RMSE), is reduced from 0.318–1.362 m to 0.273–1.032 m by using the proposed ensemble learning.

Tremor Severity and Operative Parameters Predict Imbalance in Patients Undergoing Focused Ultrasound Thalamotomy.

Imbalance is the most commonly reported side effect following focused ultrasound (FUS) thalamotomy for essential tremor (ET). It remains unknown which patients are more likely to develop imbalance following FUS treatment. To identify preoperative and treatment-related sonication parameters that are predictive of imbalance following FUS treatment. We retrospectively collected demographic data, preoperative Fahn-Tolosa-Marin Clinical Rating Scale for Tremor (FTM) scores and FUS treatment parameters in patients undergoing FUS thalamotomy for treatment of ET. The presence of imbalance was evaluated at several discrete time-points with up to 4 years of follow-up. Multiple machine learning classifiers were built and evaluated, aiming to maximize accuracy while minimizing feature set. Of the 297 patients identified, the presence of imbalance peaked at 1 week following operation at 79%. This declined rapidly with 29% reporting imbalance at 3 months, and only 15% at 4 years. At 1 week, total preoperative FTM scores and Maximum Energy delivered in FUS could predict the presence of imbalance at 92.8% accuracy. At 3 months, the total preoperative FTM scores and maximum power delivered could predict the presence of imbalance with 90.6% accuracy. Post-operative lesion size and extent into thalamic nuclei, internal capsule, and subthalamic regions were identified as likely key underlying drivers of these predictors. A machine learning model based on preoperative tremor scores and maximum energy/power delivered predicted the development of short-term imbalance and long-term imbalance following FUS thalamotomy.

Machine Learning and Remote Sensing Applications for Assessing Land Use and Land Cover Changes for Under-monitored Basin

Urban sprawling caused by industrial and economic growth has significantly affected land use and land cover (LULC). Using satellite imagery for real-time examination in Kuantan has become exceedingly expensive due to the scarcity and obsolescence of real-time LULC data. With the advent of remote sensing and geographical information systems, LULC change assessment is feasible. A quantitative assessment of image classification schemes (supervised classification using maximum likelihood and deep learning classification using random forest) was examined using 2022 Sentinel-2 satellite imagery to measure its performance. Kappa coefficient and overall accuracy were used to determine the classification accuracy. Then, 32 years of LULC changes in Kuantan were investigated using Landsat 5 TM, Landsat 8 OLI, and Sentinel-2 based on the best classifier. Random forest classification outperformed maximum likelihood classification with an overall accuracy of 85% compared to 92.8%. The findings also revealed that urbanisation is the main factor contributing to land changes in Kuantan, with a 32% increase in the build-up region and 32% in forest degradation. Despite the subtle and extremely dynamic connection between ecosystems, resources, and settlement, these LULC changes can be depicted using satellite imagery. With the precision of using a suitable classification scheme based on comprehensive, accurate and precise LULC maps can be generated, capturing the essence of spatial dynamics, especially in under-monitored basins. This study provides an overview of the current situation of LULC changes in Kuantan, along with the driving factors that can help the authorities promote sustainable development goals.

Partial Multi-Label Learning via Exploiting Instance and Label Correlations

The goal of partial multi-label learning is to induce a multi-label classifier from partial multi-label data where each instance is annotated with a number of candidate labels but only a subset of them are valid. Many of the existing studies either fail to fully utilize instance and label correlations to eliminate noisy labels or build an oversimplified multi-label classifier, both of which are unfavorable for the improvement of generalization performance. In this article, we put forward a novel model named Pml-ilc to learn a multi-label classifier from partial multi-label data. Specifically, Pml-ilc first encodes instances and labels into a compact semantic space and takes full advantage of instance and label correlations to eliminate noisy labels. Then, it induces a linear mapping from the feature space to the label space while exploiting label-specific features and instance correlations to facilitate the multi-label classifier learning process. Finally, the above two steps are combined into a joint optimization problem and an efficient alternating optimization procedure is developed to find a satisfactory solution. Extensive experiments show that Pml-ilc achieves superior performance on both real-world and synthetic partial multi-label datasets in terms of different evaluation metrics.

Dual-branch network with hypergraph feature augmentation and adaptive logits adjustment for long-tailed visual recognition

The long-tailed distribution presents issues of data scarcity and significant class imbalance, which causes the model to increase the prediction tendency for the head classes and reduce performance on tail classes. The decoupling learning strategy separates the learning process into representation learning and classifier learning to enhance model performance. However, this approach does not delve into the complex semantic relationships within images and overlooks the bias of the model confidence. To address this pervasive issue, we propose a novel algorithm called Hypergraph Feature Augmentation and Adaptive Logits Adjustment for Long-tailed Visual Recognition (HALR) based on a decoupling learning framework. For the representation learning task, we extract hypergraph features from the mixed input samples to capture the global spatial contextual semantic information of the images. For the classifier learning task, we propose an adaptive logits adjustment function that automatically corrects prediction score biases, thereby yielding robust decision boundaries. Extensive experiments on widely-used benchmark datasets, including CIFAR10/100-LT, ImageNet-LT, and iNaturalist 2018, validate the efficacy of HALR. For example, the test accuracy of HALR on CIFAR10-LT with an imbalance factor of 100 is 88.25 %, which is a 17.85 % improvement over the baseline model accuracy and 0.68 % higher than the state-of-the-art model. The results demonstrate that HALR effectively optimizes representation and classifier learning tasks for long-tailed learning, mitigating the negative impact of class imbalance and data scarcity.

Enhanced federated recognition mechanism based on spatial-temporal model with split learning for multi-view human activity classification in edge intelligent network

Enhanced federated recognition mechanism based on spatial-temporal model with split learning for multi-view human activity classification in edge intelligent network

Sample feature enhancement model based on heterogeneous graph representation learning for few-shot relation classification

Few-Shot Relation Classification (FSRC) aims to predict novel relationships by learning from limited samples. Graph Neural Network (GNN) approaches for FSRC constructs data as graphs, effectively capturing sample features through graph representation learning. However, they often face several challenges: 1) They tend to neglect the interactions between samples from different support sets and overlook the implicit noise in labels, leading to sub-optimal sample feature generation. 2) They struggle to deeply mine the diverse semantic information present in FSRC data. 3) Over-smoothing and overfitting limit the model's depth and adversely affect overall performance. To address these issues, we propose a Sample Representation Enhancement model based on Heterogeneous Graph Neural Network (SRE-HGNN) for FSRC. This method leverages inter-sample and inter-class associations (i.e., label mutual attention) to effectively fuse features and generate more expressive sample representations. Edge-heterogeneous GNNs are employed to enhance sample features by capturing heterogeneous information of varying depths through different edge attentions. Additionally, we introduce an attention-based neighbor node culling method, enabling the model to stack higher levels and extract deeper inter-sample associations, thereby improving performance. Finally, experiments are conducted for the FSRC task, and SRE-HGNN achieves an average accuracy improvement of 1.84% and 1.02% across two public datasets.

Machine Learning for Cloud Data Classification and Anomaly Intrusion Detection

Machine Learning for Cloud Data Classification and Anomaly Intrusion Detection

Animal family discrimination from hair using ATR-FTIR and machine learning methods for applications in illegal wildlife trafficking.

Wildlife forensics plays a pivotal role in the combating illegal trafficking, supporting biodiversity conservation, and aiding in the identification of animals in wildlife. Animal hair, often found in trafficking crimes, serves as vital biological evidence that can provide significant information for animal identification. This study proposes a novel method integrating machine learning classifiers with Fourier transform infrared (FTIR) spectroscopy in attenuated total reflectance (ATR) mode to enhance the effectiveness of animal identification in wildlife forensic casework. Additionally, compound microscopy has also been utilized as a preliminary tool to perform morphological analysis of hair samples from four animal families, including Bovidae, Cervidae, Elephantidae, and Felidae. Further, chemical profiling through spectral data revealed significant overlapping peaks between family Bovidae and Cervidae. The classification experiment provides the random forest (RF) classifier as the most effective for family discrimination model. This research offers valuable insights for wildlife forensics by improving the identification accuracy of unknown hair samples, thus enhancing the overall effectiveness in forensic investigations.

Optimizing Supply Chain Efficiency Using Cross-Efficiency Analysis and Inverse DEA Models

Our study employs cross-efficiency analysis (CEA) and machine learning techniques to optimize supply chain performance. By integrating inverse DEA models with directional distance functions, we measure operational efficiency across various decision-making units (DMUs), accounting for undesirable outputs such as excess costs and emissions. Our results indicate a 20% improvement in market recognition efficiency and a 15% increase in earnings persistence efficiency after model application. Additionally, machine learning classifiers, including Random Forest and Support Vector Machines, further enhanced predictive accuracy, with Random Forest achieving the lowest mean absolute error of 0.07. These findings underscore the effectiveness of advanced analytical models in improving supply chain resilience and decision-making accuracy, contributing to sustainable operational performance.

Differentiating MYCN-amplified RB1 wild-type retinoblastoma from biallelic RB1 mutant retinoblastoma using MR-based radiomics: a retrospective multicenter case–control study

MYCN-amplified RB1 wild-type (MYCNampRB1+/+) retinoblastoma is a rare and aggressive subtype, often resistant to standard therapies. Identifying unique MRI features is crucial for diagnosing this subtype, as biopsy is not recommended. This study aimed to differentiate MYCNampRB1+/+ from the most prevalent RB1-/- retinoblastoma using pretreatment MRI and radiomics. Ninety-eight unilateral retinoblastoma patients (19 MYCN cases and 79 matched controls) were included. Tumors on T2-weighted MR images were manually delineated and validated by experienced radiologists. Radiomics analysis extracted 120 features per tumor. Several combinations of feature selection methods, oversampling techniques and machine learning (ML) classifiers were evaluated in a repeated fivefold cross-validation machine learning pipeline to yield the best-performing prediction model for MYCN. The best model used univariate feature selection, data oversampling (duplicating MYCN cases), and logistic regression classifier, achieving a mean AUC of 0.78 (SD 0.12). SHAP analysis highlighted lower sphericity, higher flatness, and greater gray-level heterogeneity as predictive for MYCNampRB1+/+ status, yielding an AUC of 0.81 (SD 0.11). This study shows the potential of MRI-based radiomics to distinguish MYCNampRB1+/+ and RB1-/- retinoblastoma subtypes.

Extracting Meso- and Microscale Patterns of Urban Morphology Evolution: Evidence from Binhai New Area of Tianjin, China

Understanding and recognizing urban morphology evolution is a crucial issue in urban planning, with extensive research dedicated to detecting the extent of urban expansion. However, as urban development patterns shift from incremental expansion to stock optimization, related studies on meso- and microscale urban morphology evolution face limitations such as insufficient spatiotemporal data granularity, poor generalizability, and inability to extract internal evolution patterns. This study employs deep learning and meso-/microscopic urban form indicators to develop a generic framework for extracting and describing the evolution of meso-/microscale urban morphology. The framework includes three steps: constructing specific urban morphology datasets, semantic segmentation to extract urban form, and mapping urban form evolution using the Tile-based Urban Change (TUC) classification system. We applied this framework to conduct a combined quantitative and qualitative analysis of the internal urban morphology evolution of Binhai New Area from 2009 to 2022, with detailed visualizations of morphology evolution at each time point. The study identified that different locations in the area exhibited seven distinct evolution patterns: edge areal expansion, preservation of developmental potential, industrial land development pattern, rapid comprehensive demolition and construction pattern, linear development pattern, mixed evolution, and stable evolution. The results indicate that in the stock development phase, high-density urban areas exhibit multidimensional development characteristics by region, period, and function. Our work demonstrates the potential of using deep learning and grid classification indicators to study meso-/microscale urban morphology evolution, providing a scalable, cost-effective, quantitative, and portable approach for historical urban morphology understanding.

Comparing quantum machine learning and classical machine learning for in vitro regeneration of cowpea (Vigna unguiculata)

Comparing quantum machine learning and classical machine learning for in vitro regeneration of cowpea (Vigna unguiculata)

Archetypal Analysis of Kidney Allograft Biopsies Using Next-generation Sequencing Technology.

In kidney transplantation, molecular diagnostics may be a valuable approach to improve the precision of the diagnosis. Using next-generation sequencing (NGS), we aimed to identify clinically relevant archetypes. We conducted an Illumina bulk RNA sequencing on 770 kidney biopsies (540 kidney recipients) collected between 2006 and 2021 from 11 European centers. Differentially expressed genes were determined for 11 Banff lesions. An ElasticNet model was used for feature selection, and 4 machine learning classifiers were trained to predict the probability of presence of the lesions. NGS-based classifiers were used in an unsupervised archetypal analysis to different archetypes. The association of the archetypes with allograft survival was assessed using the iBox risk prediction score. The ElasticNet feature selection reduced the number of the genes from a range of 859-10 830 to a range of 52-867 genes. NGS-based classifiers demonstrated robust performances (precision-recall area under the curves 0.708-0.980) in predicting the Banff lesions. Archetypal analysis revealed 8 distinct phenotypes, each characterized by distinct clinical, immunological, and histological features. Although the archetypes confirmed the well-defined Banff rejection phenotypes for T cell-mediated rejection and antibody-mediated rejection, equivocal histologic antibody-mediated rejection, and borderline diagnoses were reclassified into different archetypes based on their molecular signatures. The 8 NGS-based archetypes displayed distinct allograft survival profiles with incremental graft loss rates between archetypes, ranging from 90% to 56% rates 7 y after evaluation (P < 0.0001). Using molecular phenotyping, 8 archetypes were identified. These NGS-based archetypes might improve disease characterization, reclassify ambiguous Banff diagnoses, and enable patient-specific risk stratification.

Development of an Artificial Intelligence System for Distinguishing Malignant from Benign Soft Tissue Tumors Using Contrast-Enhanced MR Images.

The integration of artificial intelligence (AI) into orthopedics has enhanced the diagnosis of various conditions; however, its use in diagnosing soft-tissue tumors remains limited owing to its complexity. This study aimed to develop and assess an AI-driven diagnostic support system for magnetic resonance imaging (MRI)-based soft-tissue tumor diagnosis, potentially improving accuracy and aiding radiologists and orthopedic surgeons. Experienced orthopedic oncologists and radiologists annotated 720 images from 77 cases (41 benign and 36 malignant soft-tissue tumors). Eleven tumor subtypes were identified and classified into benign and malignant groups based on histological diagnosis. Utilizing the standard machine learning classifier pipeline, we examined and down-selected imaging protocols and their predominant radiomic features within the tumor's three-dimensional region to differentiate between benign and malignant tumors. Among the scan protocols, contrast-enhanced T1 weighted fat-suppressed images showed the most accurate classification based on radiomics features. We focused on the two-dimensional features from the largest tumor boundary surface and its neighboring slices, leveraging texture-based radiomic and deep convolutional neural network features from a pretrained VGG19 model. The test data comprised 44 contrast-enhanced images (22 benign and 22 malignant soft-tissue tumors) containing six malignant and five benign subtypes distinct from the training data. We compared expert and non-expert human performances against AI by assessing malignancy detection and the time required for classification. The AI model showed comparable accuracy (AUC 0.91) to that of radiologists (AUC 0.83) and orthopedic surgeons (AUC 0.73). Notably, the AI model processed data approximately 400 times faster than its human counterparts, showcasing its capacity to significantly boost diagnostic efficiency. We developed an AI-driven diagnostic support system for MRI-based soft-tissue tumor diagnosis. While additional refinement is necessary for clinical applications, our system has exhibited promising potential in differentiating between benign and malignant soft-tissue tumors based on MRI.

Convergent reductive evolution in bee-associated lactic acid bacteria.

Distantly related organisms may evolve similar traits when exposed to similar environments or engaging in certain lifestyles. Several members of the Lactobacillaceae [lactic acid bacteria (LAB)] family are frequently isolated from the floral niche, mostly from bees and flowers. In some floral LAB species (henceforth referred to as bee-associated LAB), distinctive genomic (e.g., genome reduction) and phenotypic (e.g., preference for fructose over glucose or fructophily) features were recently documented. These features are found across distantly related species, raising the hypothesis that specific genomic and phenotypic traits evolved convergently during adaptation to the floral environment. To test this hypothesis, we examined representative genomes of 369 species of bee-associated and non-bee-associated LAB. Phylogenomic analysis unveiled seven independent ecological shifts toward the bee environment in LAB. In these species, we observed significant reductions of genome size, gene repertoire, and GC content. Using machine leaning, we could distinguish bee-associated from non-bee-associated species with 94% accuracy, based on the absence of genes involved in metabolism, osmotic stress, or DNA repair. Moreover, we found that the most important genes for the machine learning classifier were seemingly lost, independently, in multiple bee-associated lineages. One of these genes, acetaldehyde-alcohol dehydrogenase (adhE), encodes a bifunctional aldehyde-alcohol dehydrogenase which has been associated with the evolution of fructophily, a rare phenotypic trait that is pervasive across bee-associated LAB species. These results suggest that the independent evolution of distinctive phenotypes in bee-associated LAB has been largely driven by independent losses of the same sets of genes.IMPORTANCESeveral LAB species are intimately associated with bees and exhibit unique biochemical properties with potential for food applications and honeybee health. Using a machine learning-based approach, our study shows that adaptation of LAB to the bee environment was accompanied by a distinctive genomic trajectory deeply shaped by gene loss. Several of these gene losses occurred independently in distantly related species and are linked to some of their unique biotechnologically relevant traits, such as the preference for fructose over glucose (fructophily). This study underscores the potential of machine learning in identifying fingerprints of adaptation and detecting instances of convergent evolution. Furthermore, it sheds light onto the genomic and phenotypic particularities of bee-associated bacteria, thereby deepening the understanding of their positive impact on honeybee health.

Stacked-gait: A human gait recognition scheme based on stacked autoencoders.

Human gait recognition (HGR) is the mechanism of biometrics that authors extensively employ to recognize an individuals based on their walking traits. HGR has been prominent for the past few years due to its surveillance capability. In HGR, an individual's walking attributes are utilized for identification. HGR is considered a very effective technique for recognition but faces different problematic factors that degrade its performance. The major factors are variations in clothing, carrying, walking, etc. In this paper, a new hybrid method for the classification of HGR is designed called Stacked-Gait. The system is based on six major steps; initially, image resizing is performed to overcome computation problems. In the second step, these images are converted into grey-scale to extract better features. After that, the dataset division is performed into train and test set. In the next step, the training of the autoencoders and feature extraction of the dataset are performed using training data. In the next step, the stacking of two autoencoders is also performed. After that, the stacked encoders are employed to extract features from the test data. Finally, the feature vectors are given as input to various machine learning classifiers for final classification. The method assessment is performed using the CASIA-B dataset and achieved the accuracy of 99.90, 98.10, 97.20, 97.20, 96.70, and 100 percent on 000, 180, 360, 540, 720, and 900 angles. It is pragmatic that the system gives promising results compared to recent schemes.