Individual Classifiers Research Articles

Computational Prediction of Multiple Antigen Epitopes.

Identifying antigen epitopes is essential in medical applications, such as immunodiagnostic reagent discovery, vaccine design, and drug development. Computational approaches can complement low-throughput, time-consuming, and costly experimental determination of epitopes. Currently available prediction methods, however, have moderate success predicting epitopes, which limits their applicability. Epitope prediction is further complicated by the fact that multiple epitopes may be located on the same antigen and complete experimental data is often unavailable. Here, we introduce the antigen epitope prediction program ISPIPab that combines information from two feature-based methods and a docking-based method. We demonstrate that ISPIPab outperforms each of its individual classifiers as well as other state-of-the-art methods, including those designed specifically for epitope prediction. By combining the prediction algorithm with hierarchical clustering, we show that we can effectively capture epitopes that align with available experimental data while also revealing additional novel targets for future experimental investigations. raji@yu.edu. Supplementary data are available at Bioinformatics online.

Deep learning approaches for the detection of scar presence from cine cardiac magnetic resonance adding derived parametric images.

This work proposes a convolutional neural network (CNN) that utilizes different combinations of parametric images computed from cine cardiac magnetic resonance (CMR) images, to classify each slice for possible myocardial scar tissue presence. The CNN performance comparison in respectto expert interpretation of CMR with late gadolinium enhancement (LGE) images, used as ground truth (GT), was conducted on 206 patients (158 scar, 48 control) from Centro Cardiologico Monzino (Milan, Italy) at both slice- and patient-levels. Left ventricle dynamic features were extracted in non-enhanced cine images using parametric images based on both Fourier and monogenic signal analyses. The CNN, fed with cine images and Fourier-based parametric images, achieved an area under the ROC curve of 0.86 (accuracy 0.79, F1 0.81, sensitivity 0.9, specificity 0.65, and negative (NPV) and positive (PPV) predictive values 0.83 and 0.77, respectively), for individual slice classification. Remarkably, it exhibited 1.0 prediction accuracy (F1 0.98, sensitivity 1.0, specificity 0.9, NPV 1.0, and PPV 0.97) in patient classification as a control or pathologic. The proposed approach represents a first step towards scar detection in contrast-free CMR images. Patient-level results suggest its preliminary potential as a screening tool to guide decisions regarding LGE-CMR prescription, particularly in cases where indication is uncertain.

Unveiling anomalies: harnessing machine learning for detection and insights

The rise of Internet of Things (IoT) devices has brought about an increase in security risks, emphasizing the need for effective anomaly detection systems. Previous research introduced a dynamic voting classifier to overcome overfitting or inaccurate accuracies caused by dataset imbalance. This article introduces a new method for IoT anomaly detection that employs a hybrid voting classifier, which combines several machine learning models. To solve the overfitting and class weight issues, an adaptive voting classifier is used that adjusts weights according to the highest preference for accuracy. The developing voting system increases the effectiveness of more accurate classifiers, enhancing the group’s overall capability. A proposed combined classifier combines Logistic Regression, AdaBoost, Gradient Boosting, and Multi-Layer Perceptron models using a soft voting method. To develop and assess this method, the CIC-IoT-2023 dataset is utilized, which contains 33 types of IoT attacks across 7 categories. This process includes thorough data preprocessing and feature selection from a pool of 42 available attributes. The performance of this approach is measured against individual classifiers across binary, 8-class, and 34-class classification tasks. The results highlight the effectiveness of the hybrid model. It achieves 98.95% accuracy, 76.72% recall, and 72.01% F1-score in the 34-class problem, surpassing the performance of all individual models. For the 8-class task, the hybrid classifier attains 99.39% accuracy, 90.89% recall, and an 83.01% F1-score. This demonstrates the high potential of the hybrid approach for IoT anomaly detection.

Flexible modeling of headache frequency fluctuations in migraine with hidden Markov models.

To explore hidden Markov models (HMMs) as an approach for defining clinically meaningful headache-frequency-based groups in migraine. Monthly headache frequency in patients with migraine is known to vary over time. This variation has not been completely characterized and is not well accounted for in the classification of individuals as having chronic or episodic migraine, a diagnosis with potentially significant impacts on the individual. This study investigated variation in reported headache frequency in a migraine population and proposed a model for classifying individuals by frequency while accounting for natural variation. The American Registry for Migraine Research (ARMR) was a longitudinal multisite study of United States adults with migraine. Study participants completed quarterly questionnaires and daily headache diaries. A series of HMMs were fit to monthly headache frequency data calculated from the diary data of ARMR. Changes in monthly headache frequency tended to be small, with 47% of transitions resulting in a change of 0 or 1 day. A substantial portion (24%) of months reflected daily headache with individuals ever reporting daily headache likely to consistently report daily headache. An HMM with four states with mean monthly headache frequency emissions of 3.52 (95% Prediction Interval [PI] 0-8), 10.10 (95% PI 4-17), 20.29 (95% PI 12-28), and constant 28 days/month had the best fit of the models tested. Of sequential month-to-month headache frequency transitions, 12% were across the 15-headache days chronic migraine cutoff. Under the HMM, 38.7% of those transitions involved a change in the HMM state, and the remaining 61.3% of the time, a change in chronic migraine classification was not accompanied by a change in the HMM state. A divide between the second and third states of this model aligns most strongly with the current episodic/chronic distinction, although there is a meaningful overlap between the states that supports the need for flexibility. An HMM has appealing properties for classifying individuals according to their headache frequency while accounting for natural variation in frequency. This empirically derived model may provide an informative classification approach that is more stable than the use of a single cutoff value.

Integrated Multi-Omics Analysis of Cerebrospinal Fluid in Postoperative Delirium.

Preoperative risk biomarkers for delirium may aid in identifying high-risk patients and developing intervention therapies, which would minimize the health and economic burden of postoperative delirium. Previous studies have typically used single omics approaches to identify such biomarkers. Preoperative cerebrospinal fluid (CSF) from the Healthier Postoperative Recovery study of adults ≥ 63 years old undergoing elective major orthopedic surgery was used in a matched pair delirium case-no delirium control design. We performed metabolomics and lipidomics, which were combined with our previously reported proteomics results on the same samples. Differential expression, clustering, classification, and systems biology analyses were applied to individual and combined omics datasets. Probabilistic graph models were used to identify an integrated multi-omics interaction network, which included clusters of heterogeneous omics interactions among lipids, metabolites, and proteins. The combined multi-omics signature of 25 molecules attained an AUC of 0.96 [95% CI: 0.85-1.00], showing improvement over individual omics-based classification. We conclude that multi-omics integration of preoperative CSF identifies potential risk markers for delirium and generates new insights into the complex pathways associated with delirium. With future validation, this hypotheses-generating study may serve to build robust biomarkers for delirium and improve our understanding of its pathophysiology.

Multimodal Affective Communication Analysis: Fusing Speech Emotion and Text Sentiment Using Machine Learning

Affective communication, encompassing verbal and non-verbal cues, is crucial for understanding human interactions. This study introduces a novel framework for enhancing emotional understanding by fusing speech emotion recognition (SER) and sentiment analysis (SA). We leverage diverse features and both classical and deep learning models, including Gaussian naive Bayes (GNB), support vector machines (SVMs), random forests (RFs), multilayer perceptron (MLP), and a 1D convolutional neural network (1D-CNN), to accurately discern and categorize emotions in speech. We further extract text sentiment from speech-to-text conversion, analyzing it using pre-trained models like bidirectional encoder representations from transformers (BERT), generative pre-trained transformer 2 (GPT-2), and logistic regression (LR). To improve individual model performance for both SER and SA, we employ an extended dynamic Bayesian mixture model (DBMM) ensemble classifier. Our most significant contribution is the development of a novel two-layered DBMM (2L-DBMM) for multimodal fusion. This model effectively integrates speech emotion and text sentiment, enabling the classification of more nuanced, second-level emotional states. Evaluating our framework on the EmoUERJ (Portuguese) and ESD (English) datasets, the extended DBMM achieves accuracy rates of 96% and 98% for SER, 85% and 95% for SA, and 96% and 98% for combined emotion classification using the 2L-DBMM, respectively. Our findings demonstrate the superior performance of the extended DBMM for individual modalities compared to individual classifiers and the 2L-DBMM for merging different modalities, highlighting the value of ensemble methods and multimodal fusion in affective communication analysis. The results underscore the potential of our approach in enhancing emotional understanding with broad applications in fields like mental health assessment, human–robot interaction, and cross-cultural communication.

Instability of high polygenic risk classification and mitigation by integrative scoring.

Polygenic risk scores (PRS) continue to improve with novel methods and expanding genome-wide association studies. Healthcare and third-party laboratories are increasingly deploying PRS reports to patients. Although new PRS show improving strengths of association with traits, it is unknown how the classification of high polygenic risk changes across individual PRS for the same trait. Here, we determined classification of high genetic risk from all cataloged PRS for three complex traits. While each PRS for each trait demonstrated generally consistent population-level strengths of associations, classification of individuals in the top 10% of each PRS distribution varied widely. Using the PRSMix framework, which incorporates information across several PRS to improve prediction, we generated sequential add-one-in (AOI) PRSMix_AOI scores based on order of publication. PRSMix_AOI n led to improved PRS performance and more consistent high-risk classification compared with the PRS n . The PRSMix_AOI approach provides more stable and reliable classification of high-risk as new PRS continue to be generated toward PRS standardization.

Online aggression detection using ensemble techniques on fine-tuned transformer-based language models

The widespread adoption of Online Platforms for our day-to-day life is increasingly contributing to the rise of Online Aggression and its escalation. Consequently, there is a need for a robust mechanism that could automatically recognize online aggressive content to foster a safer and more respectful online environment. In this work, we explored the usage of four well-known Transformer-based Language Models i.e. BERT, RoBERTa, ALBERT, and XLNet, and on top employed three varied kinds of Ensemble Techniques such as Logistic Regression, Random Forest, and XGBoost on the predictions obtained from those standalone Transformer Models and aim to actively contribute to the advancement of automated Cyber Aggression Identification system. We identified the noise involved and a few outright erroneous data labels in the popular shared task Aggression dataset TRAC-1 and developed an improvised dataset called ReLbTRAC-1 by resolving the identified issues. The proposed models are evaluated on improvised TRAC-1, and TRAC-2 datasets, and they outperformed the state-of-art methods on the TRAC-1 dataset and most existing approaches on the TRAC-2 dataset, including all individual Transformer-based classifiers employed in the proposed Ensemble techniques, obtaining admirable results on all metrics considered (such as Accuracy of 0.6305, and 0.7550 and weighted F1-Score of 0.6518, and 0.7512 on TRAC-1 and TRAC-2 datasets respectively).

Individual tree detection and classification from RGB satellite imagery with applications to wildfire fuel mapping and exposure assessments

Individual tree detection and classification from RGB satellite imagery with applications to wildfire fuel mapping and exposure assessments

A hybrid model for the classification of Autism Spectrum Disorder using Mu rhythm in EEG.

Autism Spectrum Disorder (ASD) is a condition with social interaction, communication, and behavioral difficulties. Diagnostic methods mostly rely on subjective evaluations and can lack objectivity. In this research Machine learning (ML) and deep learning (DL) techniques are used to enhance ASD classification. This study focuses on improving ASD and TD classification accuracy with a minimal number of EEG channels. ML and DL models are used with EEG data, including Mu Rhythm from the Sensory Motor Cortex (SMC) for classification. Non-linear features in time and frequency domains are extracted and ML models are applied for classification. The EEG 1D data is transformed into images using Independent Component Analysis-Second Order Blind Identification (ICA-SOBI), Spectrogram, and Continuous Wavelet Transform (CWT). Stacking Classifier employed with non-linear features yields precision, recall, F1-score, and accuracy rates of 78%, 79%, 78%, and 78% respectively. Including entropy and fuzzy entropy features further improves accuracy to 81.4%. In addition, DL models, employing SOBI, CWT, and spectrogram plots, achieve precision, recall, F1-score, and accuracy of 75%, 75%, 74%, and 75% respectively. The hybrid model, which combined deep learning features from spectrogram and CWT with machine learning, exhibits prominent improvement, attained precision, recall, F1-score, and accuracy of 94%, 94%, 94%, and 94% respectively. Incorporating entropy and fuzzy entropy features further improved the accuracy to 96.9%. This study underscores the potential of ML and DL techniques in improving the classification of ASD and TD individuals, particularly when utilizing a minimal set of EEG channels.

Predicting Employee Promotion using Machine Learning

Training and development are key components of professional development for people to improve their capacity. Professional development programs are typically organized around personal information like as background, personal goals, and work experience, as well as corporate objectives and job requirements. Individual employee classification is required to promote tailored training in the professional development process. As a result, this study provides a classification approach for employee classification in order to facilitate tailored training in enterprises. Machine learning methods such as Decision Tree, Random Forest, and Support Vector Machine are investigated. To cope with imbalance data, the Synthetic Minority Oversampling Technique (SMOTE) approach is applied. In this work, the open data form kaggle is used. The training and testing data are combined to generate the data for technique validation. There are three gropes: 80:20, 70:30, and 60:40. According to the classification results, the SMOTE can increase classification performance for all classifiers. Furthermore, random forest has the highest categorization accuracy.

Resolving autism spectrum disorder (ASD) through brain topologies using fMRI dataset with multi-layer perceptron (MLP)

Autism is a neurodevelopmental disorder that manifests in individuals during childhood and has enduring consequences for their social interactions and communication. The prediction of Autism Spectrum Disorder (ASD) in individuals based on the differences in brain networks and activities have been studied extensively in the recent past, however, with lower accuracies. Therefore in this research, identification at the early stage through computer-aided algorithms to differentiate between ASD and TD patients is proposed. In order to identify features, a Multi-Layer Perceptron (MLP) model is developed which utilizes logistic regression on characteristics extracted from connectivity matrices of subjects derived from fMRI images. The features that significantly contribute to the classification of individuals as having Autism Spectrum Disorder (ASD) or typically developing (TD) are identified by the logistic regression model. To enhance emphasis on essential attributes, an AND operation is integrated. This involves selecting features demonstrating statistical significance across diverse logistic regression analyses conducted on various random distributions. The iterative approach contributes to a comprehensive understanding of relevant features for accurate classification. By implementing this methodology, the estimation of feature importance became more dependable, and the potential for overfitting is moderated through the evaluation of model performance on various subsets of data. It is observed from the experimentation that the highly correlated Left Lateral Occipital Cortex and Right Lateral Occipital Cortex ROIs are only found in ASD. Also, it is noticed that the highly correlated Left Cerebellum Tonsil and Right Cerebellum Tonsil are only found in TD participants. Among the MLP classifier, a recall of 82.61 % is achieved followed by Logistic Regression with an accuracy of 72.46 %. MLP also stands out with a commendable accuracy of 83.57 % and AUC of 0.978.

Prostate cancer risk assessment and avoidance of prostate biopsies using fully automatic deep learning in prostate MRI: comparison to PI-RADS and integration with clinical data in nomograms.

Risk calculators (RCs) improve patient selection for prostate biopsy with clinical/demographic information, recently with prostate MRI using the prostate imaging reporting and data system (PI-RADS). Fully-automated deep learning (DL) analyzes MRI data independently, and has been shown to be on par with clinical radiologists, but has yet to be incorporated into RCs. The goal of this study is to re-assess the diagnostic quality of RCs, the impact of replacing PI-RADS with DL predictions, and potential performance gains by adding DL besides PI-RADS. One thousand six hundred twenty-seven consecutive examinations from 2014 to 2021 were included in this retrospective single-center study, including 517 exams withheld for RC testing. Board-certified radiologists assessed PI-RADS during clinical routine, then systematic and MRI/Ultrasound-fusion biopsies provided histopathological ground truth for significant prostate cancer (sPC). nnUNet-based DL ensembles were trained on biparametric MRI predicting the presence of sPC lesions (UNet-probability) and a PI-RADS-analogous five-point scale (UNet-Likert). Previously published RCs were validated as is; with PI-RADS substituted by UNet-Likert (UNet-Likert-substituted RC); and with both UNet-probability and PI-RADS (UNet-probability-extended RC). Together with a newly fitted RC using clinical data, PI-RADS and UNet-probability, existing RCs were compared by receiver-operating characteristics, calibration, and decision-curve analysis. Diagnostic performance remained stable for UNet-Likert-substituted RCs. DL contained complementary diagnostic information to PI-RADS. The newly-fitted RC spared 49% [252/517] of biopsies while maintaining the negative predictive value (94%), compared to PI-RADS ≥ 4 cut-off which spared 37% [190/517] (p < 0.001). Incorporating DL as an independent diagnostic marker for RCs can improve patient stratification before biopsy, as there is complementary information in DL features and clinical PI-RADS assessment. For patients with positive prostate screening results, a comprehensive diagnostic workup, including prostate MRI, DL analysis, and individual classification using nomograms can identify patients with minimal prostate cancer risk, as they benefit less from the more invasive biopsy procedure. The current MRI-based nomograms result in many negative prostate biopsies. The addition of DL to nomograms with clinical data and PI-RADS improves patient stratification before biopsy. Fully automatic DL can be substituted for PI-RADS without sacrificing the quality of nomogram predictions. Prostate nomograms show cancer detection ability comparable to previous validation studies while being suitable for the addition of DL analysis.

An automated ensemble approach using Harris Hawk optimization for visually evoked EEG signal classification.

Steady-state visually evoked potential is one of the active explorations in the brain-computer interface research. Electroencephalogram based brain computer interface studies have been widely applied to perceive solutions for real-world problems in the healthcare domain. The classification of externally bestowed visual stimuli of different frequencies on a human was experimented to identify the need of paralytic people. Although many classifiers are at the fingertip of machine learning technology, recent research has proven that ensemble learning is more efficacious than individual classifiers. Despite its efficiency, ensemble learning technology exhibits certain drawbacks like taking more time on selecting the optimal classifier subset. This research article utilizes the Harris Hawk Optimization algorithm to select the best classifier subset from the given set of classifiers. The objective of the research is to develop an efficient multi-classifier model for electroencephalogram signal classification. The proposed model utilizes the Boruta Feature Selection algorithm to select the prominent features for classification. Thus selected prominent features are fed into the multi-classifier subset which has been generated by the Harris Hawk Optimization algorithm. The results of the multi-classifier ensemble model are aggregated using Stacking, Bagging, Boosting, and Voting. The proposed model is evaluated against the acquired dataset and produces a promising accuracy of 96.1%, 98.7%, 91.91%, and 99.01% with the ensemble techniques respectively. The proposed model is also validated with other performance metrics such as sensitivity, specificity, and F1-Score. The experimental results show that the proposed model proves its supremacy in segregating the multi-class classification problem with high accuracy.

Spatial disaggregation of a legacy soil map to support digital soil and land evaluation assessments in Scotland

In recent years, the importance of soils and soil functions has been recognised for supporting the delivery of ecosystem services and for the realisation of international initiatives, such as the UN Sustainable Development Goals. At the same time, Digital Soil Mapping (DSM) has emerged as a modelling technique that can satisfy increased end-user needs for new soil datasets by producing fine resolution soils and soil property maps to support complex digital soil and land evaluation assessments. Spatial disaggregation is a popular DSM technique that is used to transform legacy soil maps to more spatially-explicit soils datasets, which can also be used in conjunction with soil databases to generate digital soil property maps. In this study, we performed spatial disaggregation of the National Soil Map of Scotland (originally published at 1:250,000 scale) at the taxonomic level of Soil Series, with the specific objective to facilitate the production of harmonised digital soil property maps to support soil and land evaluation assessments in Scotland through linking to the Scottish Soil Database. We divided Scotland into Landscape Units of similar soil and landform characteristics and trained probability random forest models within each Landscape Unit using area-proportion random sampling of both single- and multiple- (complex) Soil Series map units and selected environmental covariates to produce Soil Series probability layers at 50 m grid resolution. The performance of the disaggregated Soil Series maps was evaluated using prediction uncertainties of individual soil types and independent soil profile classifications. Evaluation results indicated that the random forest algorithm was successful in promoting effective spatial disaggregation of both single soil and complex soil polygons and provided good prediction accuracies for most soil types with the exception of some of the least extensive soil types typically found within complex map units. This was attributed mainly to algorithm's tendency to favour dominant, more extensive classes, along with its difficulty to distinguish between similar soils within spatially diverse areas. However, training Soil Series models at a Landscape Unit level instead of nationally helped to limit both the underestimation of these minority soil types and the overestimation of the dominant ones. In addition, map evaluation results showed the usefulness of using the generated conditional Soil Series probabilities for exploring soil spatial variability, especially within complex areas such as river floodplains covered by multiple alluvial and non-alluvial soils. Overall, this study demonstrates the potential of using spatial disaggregation to extract pedological knowledge embedded in legacy soil maps and use it to generate new dynamic and harmonised soil and soil property maps by effectively using readily-available and easily-updated soils information from existing databases.

Improved source classification and performance analysis using Gaia DR3

The Discrete Source Classifier (DSC) provides probabilistic classification of sources in Gaia Data Release 3 (GDR3) using a Bayesian framework and a global prior. The DSC Combmod classifier in GDR3 achieved for the extragalactic classes (quasars and galaxies) a high completeness of 92%, but a low purity of 22% (all sky, all magnitudes) due to contamination from the far larger star class. However, these single metrics mask significant variation in performance with magnitude and sky position. Furthermore, a better combination of the individual classifiers that comprise Combmod is possible. Here we compute two-dimensional (2D) representations of the completeness and the purity as a function of Galactic latitude and source brightness, and also exclude the Magellanic Clouds where stellar contamination significantly reduces the purity. Reevaluated on a cleaner validation set and without introducing changes to the published GDR3 DSC probabilities themselves, we here achieve for Combmod average 2D completenesses of 92% and 95% and average 2D purities of 55% and 89% for the quasar and galaxy classes, respectively. Since the relative proportions of extragalactic objects to stars in Gaia is expected to vary significantly with brightness and latitude, we then introduce a new prior that is a continuous function of brightness and latitude, and compute new class probabilities from the GDR3 DSC component classifiers, Specmod and Allosmod. Contrary to expectations, this variable prior only improves the performance by a few percentage points, mostly at the faint end. Significant improvement, however, is obtained by a new additive combination of Specmod and Allosmod. This classifier, Combmod-α, achieves average 2D completenesses of 82% and 93% and average 2D purities of 79% and 93% for the quasar and galaxy classes, respectively, when using the global prior. Thus, we achieve a significant improvement in purity for a small loss of completeness. The improvement is most significant for faint quasars (G≥20) where the purity rises from 20% to 62%.

Anomaly-based error and intrusion detection in tabular data: No DNN outperforms tree-based classifiers

Recent years have seen a growing involvement of researchers and practitioners in crafting Deep Neural Networks (DNNs) that seem to outperform existing machine learning approaches for solving classification problems as anomaly-based error and intrusion detection. Undoubtedly, classifiers may be very diverse among themselves, and choosing one or another is typically due to the specific task and target system. Designing and training the optimal tabular data classifier requires extensive experimentation, sensitivity analyses, big datasets, and domain-specific knowledge that may not be available at will or considered a non-strategical asset by many companies and stakeholders. This paper compares, using a total of 23 public datasets: i) traditional (tree-based, statistical) supervised classifiers, ii) DNNs that are specifically designed for classifying tabular data, iii) DNNs for image classification that are applied to tabular data after converting data points into images, alone and as ensembles. Experimental results and related discussions show clear advantages in adopting tree-based classifiers for anomaly-based error and intrusion detection in tabular data as they outperform their competitors, including DNNs. Then, individual classifiers are compared against ensembles using different combinations of the classifiers considered in this study as base-learners, providing a unified final response through many meta-learning strategies. Results show that there is no benefit in building ensembles instead of using a tree-based classifier as Random Forests, eXtreme Gradient Boosting or Extra Trees. The paper concludes that anomaly-based error and intrusion detectors for critical systems should use the old (but gold) tree-based classifiers, which are also easier to fine-tune, and understand; plus, they require less time and resources to learn their model.

Using machine learning algorithms to cluster and classify stone pine (Pinus pinea L.) populations based on seed and seedling characteristics

AbstractThe phenotype of a woody plant represents its unique morphological properties. Population discrimination and individual classification are crucial for breeding populations and conserving genetic diversity. Machine Learning (ML) algorithms are gaining traction as powerful tools for predicting phenotypes. The present study is focused on classifying and clustering the seeds and seedlings in terms of morphological characteristics using ML algorithms. In addition, the k-means algorithm is used to determine the ideal number of clusters. The results obtained from the k-means algorithm were then compared with reality. The best classification performance achieved by the Random Forest algorithm was an accuracy of 0.648 and an F1-Score of 0.658 for the seed traits. Also, the best classification performance for stone pine seedlings was observed for the k-Nearest Neighbors algorithm (k = 18), for which the accuracy and F1-Score were 0.571 and 0.582, respectively. The best clustering performance was achieved with k = 2 for the seed (average Silhouette index = 0.48) and seedling (average Silhouette Index = 0.51) traits. According to the principal component analysis, two dimensions accounted for 97% and 63% of the traits of seeds and seedlings, respectively. The most important features between the seed and seedling traits were cone weight and bud set, respectively. This study will provide a foundation and motivation for future efforts in forest management practices, particularly regarding reforestation, yield optimization, and breeding programs.

A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME

eXplainable artificial intelligence (XAI) methods have emerged to convert the black box of machine learning (ML) models into a more digestible form. These methods help to communicate how the model works with the aim of making ML models more transparent and increasing the trust of end‐users in their output. SHapley Additive exPlanations (SHAP) and Local Interpretable Model Agnostic Explanation (LIME) are two widely used XAI methods, particularly with tabular data. In this perspective piece, the way the explainability metrics of these two methods are generated is discussed and a framework for the interpretation of their outputs, highlighting their weaknesses and strengths is proposed. Specifically, their outcomes in terms of model‐dependency and in the presence of collinearity among the features, relying on a case study from the biomedical domain (classification of individuals with or without myocardial infarction) are discussed. The results indicate that SHAP and LIME are highly affected by the adopted ML model and feature collinearity, raising a note of caution on their usage and interpretation.

An efficient ensemble learning method based on multi-objective feature selection

Ensemble learning (EL) boosts model prediction performance across various domains through two main steps: generating individual classifiers (ICs) and combining them. Creating accurate and diverse ICs is crucial for a strong ensemble, while selecting the best ICs, known as ensemble selection (ES), is critical yet challenging due to the accuracy-diversity trade-off and the lack of agreed-upon diversity metrics. This paper introduces an EL strategy that uses multi-objective feature selection (MOFS) and a feature relevance-guided selection to tackle these challenges. Our approach uses a hybrid MOFS algorithm to produce accurate and diverse ICs, and then it employs a novel knowledge-based feature-relevance-guided metric for precise diversity assessment during ES. The ES issue is cast as an optimization problem, aiming to maximize both diversity and accuracy, and an efficient ES algorithm is developed to select optimal ICs. Extensive tests on public datasets and a real-world prediction task demonstrate the effectiveness of our method, especially in achieving high accuracy.