Mach. Learn. Knowl. Extr., Volume 5, Issue 4 (December 2023) – 27 articles

The rapid development of semi-supervised machine learning (SSML) algorithms has shown enhanced versatility, but pinpointing the primary influencing factors remains a challenge. Historically, deep neural networks (DNNs) have been used to underpin these algorithms, resulting in improved classification precision. This study aims to delve into the performance determinants of SSML models by employing post-hoc explainable artificial intelligence (XAI) methods. By analyzing the components of well-established SSML algorithms and comparing them to newer counterparts, this work redefines semi-supervised computation processes for both data preprocessing and classification. Integrating different types of DNNs, we evaluated the effects of parameter adjustments during training across varied labeled and unlabeled data proportions. Our analysis of 45 experiments showed a notable 8% drop in training loss and a 6.75% enhancement in learning precision when using the Shake-Shake26 classifier with the RemixMatch SSML algorithm. Additionally, our findings suggest a strong positive relationship between the amount of labeled data and training duration, indicating that more labeled data leads to extended training periods, which further influences parameter adjustments in learning processes. Full article

The proliferation of Android-based devices has brought about an unprecedented surge in mobile application usage, making the Android ecosystem a prime target for cybercriminals. In this paper, a new method for Android malware classification is proposed. The method implements a convolutional neural network for malware classification using images. The research presents a novel approach to transforming the Android Application Package (APK) into a grayscale image. The image creation utilizes natural language processing techniques for text cleaning, extraction, and fuzzy hashing to represent the decompiled code from the APK in a set of hashes after preprocessing, where the image is composed of n fuzzy hashes that represent an APK. The method was tested on an Android malware dataset with 15,493 samples of five malware types. The proposed method showed an increase in accuracy compared to others in the literature, achieving up to 98.24% in the classification task. Full article

Deep Learning has enabled significant progress towards more accurate predictions and is increasingly integrated into our everyday lives in real-world applications; this is true especially for Convolutional Neural Networks (CNNs) in the field of image analysis. Nevertheless, it has been shown that Deep Learning is vulnerable against well-crafted, small perturbations to the input, i.e., adversarial examples. Defending against such attacks is therefore crucial to ensure the proper functioning of these models—especially when autonomous decisions are taken in safety-critical applications, such as autonomous vehicles. In this work, shallow machine learning models, such as Logistic Regression and Support Vector Machine, are utilised as surrogates of a CNN based on the assumption that they would be differently affected by the minute modifications crafted for CNNs. We develop three detection strategies for adversarial examples by analysing differences in the prediction of the surrogate and the CNN model: namely, deviation in (i) the prediction, (ii) the distance of the predictions, and (iii) the confidence of the predictions. We consider three different feature spaces: raw images, extracted features, and the activations of the CNN model. Our evaluation shows that our methods achieve state-of-the-art performance compared to other approaches, such as Feature Squeezing, MagNet, PixelDefend, and Subset Scanning, on the MNIST, Fashion-MNIST, and CIFAR-10 datasets while being robust in the sense that they do not entirely fail against selected single attacks. Further, we evaluate our defence against an adaptive attacker in a grey-box setting. Full article

We propose and implement an interpretable machine learning classification model for Explainable AI (XAI) based on expressive Boolean formulas. Potential applications include credit scoring and diagnosis of medical conditions. The Boolean formula defines a rule with tunable complexity (or interpretability) according to which input data are classified. Such a formula can include any operator that can be applied to one or more Boolean variables, thus providing higher expressivity compared to more rigid rule- and tree-based approaches. The classifier is trained using native local optimization techniques, efficiently searching the space of feasible formulas. Shallow rules can be determined by fast Integer Linear Programming (ILP) or Quadratic Unconstrained Binary Optimization (QUBO) solvers, potentially powered by special-purpose hardware or quantum devices. We combine the expressivity and efficiency of the native local optimizer with the fast operation of these devices by executing non-local moves that optimize over the subtrees of the full Boolean formula. We provide extensive numerical benchmarking results featuring several baselines on well-known public datasets. Based on the results, we find that the native local rule classifier is generally competitive with the other classifiers. The addition of non-local moves achieves similar results with fewer iterations. Therefore, using specialized or quantum hardware could lead to a significant speedup through the rapid proposal of non-local moves. Full article

In this paper, we present a comparative study of modern semantic segmentation loss functions and their resultant impact when applied with state-of-the-art off-road datasets. Class imbalance, inherent in these datasets, presents a significant challenge to off-road terrain semantic segmentation systems. With numerous environment classes being extremely sparse and underrepresented, model training becomes inefficient and struggles to comprehend the infrequent minority classes. As a solution to this problem, loss functions have been configured to take class imbalance into account and counteract this issue. To this end, we present a novel loss function, Focal Class-based Intersection over Union (FCIoU), which directly targets performance imbalance through the optimization of class-based Intersection over Union (IoU). The new loss function results in a general increase in class-based performance when compared to state-of-the-art targeted loss functions. Full article

The cryptocurrency market has grown significantly, and this quick growth has given rise to scams. It is necessary to put fraud detection mechanisms in place. The challenge of inadequate labeling is addressed in this work, which is a barrier to the training of high-performance supervised classifiers. It aims to lessen the necessity for laborious and time-consuming manual labeling. Some unlabeled data points have labels that are more pertinent and informative for the supervised model to learn from. The viability of utilizing unsupervised anomaly detection algorithms and active learning strategies to build an iterative process of acquiring labeled transactions in a cold start scenario, where there are no initial-labeled transactions, is being investigated. Investigating anomaly detection capabilities for a subset of data that maximizes supervised models’ learning potential is the goal. The anomaly detection algorithms under performed, according to the results. The findings underscore the need that anomaly detection algorithms be reserved for situations involving cold starts. As a result, using active learning techniques would produce better outcomes and supervised machine learning model performance. Full article

YOLO has become a central real-time object detection system for robotics, driverless cars, and video monitoring applications. We present a comprehensive analysis of YOLO’s evolution, examining the innovations and contributions in each iteration from the original YOLO up to YOLOv8, YOLO-NAS, and YOLO with transformers. We start by describing the standard metrics and postprocessing; then, we discuss the major changes in network architecture and training tricks for each model. Finally, we summarize the essential lessons from YOLO’s development and provide a perspective on its future, highlighting potential research directions to enhance real-time object detection systems. Full article

In an era characterised by rapid technological advancement, the application of algorithmic approaches to address complex problems has become crucial across various disciplines. Within the realm of education, there is growing recognition of the pivotal role played by computational thinking (CT). This skill set has emerged as indispensable in our ever-evolving digital landscape, accompanied by an equal need for effective methods to assess and measure these skills. This research places its focus on the Cross Array Task (CAT), an educational activity designed within the Swiss educational system to assess students’ algorithmic skills. Its primary objective is to evaluate pupils’ ability to deconstruct complex problems into manageable steps and systematically formulate sequential strategies. The CAT has proven its effectiveness as an educational tool in tracking and monitoring the development of CT skills throughout compulsory education. Additionally, this task presents an enthralling avenue for algorithmic research, owing to its inherent complexity and the necessity to scrutinise the intricate interplay between different strategies and the structural aspects of this activity. This task, deeply rooted in logical reasoning and intricate problem solving, often poses a substantial challenge for human solvers striving for optimal solutions. Consequently, the exploration of computational power to unearth optimal solutions or uncover less intuitive strategies presents a captivating and promising endeavour. This paper explores two distinct algorithmic approaches to the CAT problem. The first approach combines clustering, random search, and move selection to find optimal solutions. The second approach employs reinforcement learning techniques focusing on the Proximal Policy Optimization (PPO) model. The findings of this research hold the potential to deepen our understanding of how machines can effectively tackle complex challenges like the CAT problem but also have broad implications, particularly in educational contexts, where these approaches can be seamlessly integrated into existing tools as a tutoring mechanism, offering assistance to students encountering difficulties. This can ultimately enhance students’ CT and problem-solving abilities, leading to an enriched educational experience. Full article

Human Pose Estimation (HPE) is the task that aims to predict the location of human joints from images and videos. This task is used in many applications, such as sports analysis and surveillance systems. Recently, several studies have embraced deep learning to enhance the performance of HPE tasks. However, building an efficient HPE model is difficult; many challenges, like crowded scenes and occlusion, must be handled. This paper followed a systematic procedure to review different HPE models comprehensively. About 100 articles published since 2014 on HPE using deep learning were selected using several selection criteria. Both image and video data types of methods were investigated. Furthermore, both single and multiple HPE methods were reviewed. In addition, the available datasets, different loss functions used in HPE, and pretrained feature extraction models were all covered. Our analysis revealed that Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are the most used in HPE. Moreover, occlusion and crowd scenes remain the main problems affecting models’ performance. Therefore, the paper presented various solutions to address these issues. Finally, this paper highlighted the potential opportunities for future work in this task. Full article

The perception system is a safety-critical component that directly impacts the overall safety of autonomous driving systems (ADSs). It is imperative to ensure the robustness of the deep-learning model used in the perception system. However, studies have shown that these models are highly vulnerable to the adversarial perturbation of input data. The existing works mainly focused on studying the impact of these adversarial attacks on classification rather than regression models. Therefore, this paper first introduces two generalized methods for perturbation-based attacks: (1) We used naturally occurring noises to create perturbations in the input data. (2) We introduce a modified square, HopSkipJump, and decision-based/boundary attack to attack the regression models used in ADSs. Then, we propose a deep-autoencoder-based adversarial attack detector. In addition to offline evaluation metrics (e.g., F1 score and precision, etc.), we introduce an online evaluation framework to evaluate the robustness of the model under attack. The framework considers the reconstruction loss of the deep autoencoder that validates the robustness of the models under attack in an end-to-end fashion at runtime. Our experimental results showed that the proposed adversarial attack detector could detect square, HopSkipJump, and decision-based/boundary attacks with a true positive rate (TPR) of 93%. Full article

Verbal autopsies (VA) are commonly used in Low- and Medium-Income Countries (LMIC) to determine cause of death (CoD) where death occurs outside clinical settings, with the most commonly used international gold standard being physician medical certification. Interviewers elicit information from relatives of the deceased, regarding circumstances and events that might have led to death. This information is stored in textual format as VA narratives. The narratives entail detailed information that can be used to determine CoD. However, this approach still remains a manual task that is costly, inconsistent, time-consuming and subjective (prone to errors), amongst many drawbacks. As such, this negatively affects the VA reporting process, despite it being vital for strengthening health priorities and informing civil registration systems. Therefore, this study seeks to close this gap by applying novel deep learning (DL) interpretable approaches for reviewing VA narratives and generate CoD prediction in a timely, easily interpretable, cost-effective and error-free way. We validate our DL models using optimisation and performance accuracy machine learning (ML) curves as a function of training samples. We report on validation with training set accuracy (LSTM = 76.11%, CNN = 76.35%, and SEDL = 82.1%), validation accuracy (LSTM = 67.05%, CNN = 66.16%, and SEDL = 82%) and test set accuracy (LSTM = 67%, CNN = 66.2%, and SEDL = 82%) for our models. Furthermore, we also present Local Interpretable Model-agnostic Explanations (LIME) for ease of interpretability of the results, thereby building trust in the use of machines in healthcare. We presented robust deep learning methods to determine CoD from VAs, with the stacked ensemble deep learning (SEDL) approaches performing optimally and better than Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN). Our empirical results suggest that ensemble DL methods may be integrated in the CoD process to help experts get to a diagnosis. Ultimately, this will reduce the turnaround time needed by physicians to go through the narratives in order to be able to give an appropriate diagnosis, cut costs and minimise errors. This study was limited by the number of samples needed for training our models and the high levels of lexical variability in the words used in our textual information. Full article

Machine learning (ML) has become a critical technology in the defense sector, enabling the development of advanced systems for threat detection, decision making, and autonomous operations. However, the increasing ML use in defense systems has raised ethical concerns related to accountability, transparency, and bias. In this paper, we provide a comprehensive analysis of the impact of ML on the defense sector, including the benefits and drawbacks of using ML in various applications such as surveillance, target identification, and autonomous weapons systems. We also discuss the ethical implications of using ML in defense, focusing on privacy, accountability, and bias issues. Finally, we present recommendations for mitigating these ethical concerns, including increased transparency, accountability, and stakeholder involvement in designing and deploying ML systems in the defense sector. Full article

Implantable Cardiac Monitor (ICM) devices are demonstrating, as of today, the fastest-growing market for implantable cardiac devices. As such, they are becoming increasingly common in patients for measuring heart electrical activity. ICMs constantly monitor and record a patient’s heart rhythm, and when triggered, send it to a secure server where health care professionals (HCPs) can review it. These devices employ a relatively simplistic rule-based algorithm (due to energy consumption constraints) to make alerts for abnormal heart rhythms. This algorithm is usually parameterized to an over-sensitive mode in order to not miss a case (resulting in a relatively high false-positive rate), and this, combined with the device’s nature of constantly monitoring the heart rhythm and its growing popularity, results in HCPs having to analyze and diagnose an increasingly growing number of data. In order to reduce the load on the latter, automated methods for ECG analysis are nowadays becoming a great tool to assist HCPs in their analysis. While state-of-the-art algorithms are data-driven rather than rule-based, training data for ICMs often consist of specific characteristics that make their analysis unique and particularly challenging. This study presents the challenges and solutions in automatically analyzing ICM data and introduces a method for its classification that outperforms existing methods on such data. It carries this out by combining high-frequency noise detection (which often occurs in ICM data) with a semi-supervised learning pipeline that allows for the re-labeling of training episodes and by using segmentation and dimension-reduction techniques that are robust to morphology variations of the sECG signal (which are typical to ICM data). As a result, it performs better than state-of-the-art techniques on such data with, e.g., an F1 score of 0.51 vs. 0.38 of our baseline state-of-the-art technique in correctly calling atrial fibrillation in ICM data. As such, it could be used in numerous ways, such as aiding HCPs in the analysis of ECGs originating from ICMs by, e.g., suggesting a rhythm type. Full article

The rise of machine-learning applications in domains with critical end-user impact has led to a growing concern about the fairness of learned models, with the goal of avoiding biases that negatively impact specific demographic groups. Most existing bias-mitigation strategies adapt the importance of data instances during pre-processing. Since fairness is a contextual concept, we advocate for an interactive machine-learning approach that enables users to provide iterative feedback for model adaptation. Specifically, we propose to adapt the explanatory interactive machine-learning approach Caipi for fair machine learning. FairCaipi incorporates human feedback in the loop on predictions and explanations to improve the fairness of the model. Experimental results demonstrate that FairCaipi outperforms a state-of-the-art pre-processing bias mitigation strategy in terms of the fairness and the predictive performance of the resulting machine-learning model. We show that FairCaipi can both uncover and reduce bias in machine-learning models and allows us to detect human bias. Full article

Human capability to perform routine tasks declines with age and age-related problems. Remote human activity recognition (HAR) is beneficial for regular monitoring of the elderly population. This paper addresses the problem of the continuous detection of daily human activities using a mm-wave Doppler radar. In this study, two strategies have been employed: the first method uses un-equalized series of activities, whereas the second method utilizes a gradient-based strategy for equalization of the series of activities. The dynamic time warping (DTW) algorithm and Long Short-term Memory (LSTM) techniques have been implemented for the classification of un-equalized and equalized series of activities, respectively. The input for DTW was provided using three strategies. The first approach uses the pixel-level data of frames (UnSup-PLevel). In the other two strategies, a convolutional variational autoencoder (CVAE) is used to extract Un-Supervised Encoded features (UnSup-EnLevel) and Supervised Encoded features (Sup-EnLevel) from the series of Doppler frames. The second approach for equalized data series involves the application of four distinct feature extraction methods: i.e., convolutional neural networks (CNN), supervised and unsupervised CVAE, and principal component Analysis (PCA). The extracted features were considered as an input to the LSTM. This paper presents a comparative analysis of a novel supervised feature extraction pipeline, employing Sup-ENLevel-DTW and Sup-EnLevel-LSTM, against several state-of-the-art unsupervised methods, including UnSUp-EnLevel-DTW, UnSup-EnLevel-LSTM, CNN-LSTM, and PCA-LSTM. The results demonstrate the superiority of the Sup-EnLevel-LSTM strategy. However, the UnSup-PLevel strategy worked surprisingly well without using annotations and frame equalization. Full article

The inherent dependency of deep learning models on labeled data is a well-known problem and one of the barriers that slows down the integration of such methods into different fields of applied sciences and engineering, in which experimental and numerical methods can easily generate a colossal amount of unlabeled data. This paper proposes an unsupervised domain adaptation methodology that mimics the peer review process to label new observations in a different domain from the training set. The approach evaluates the validity of a hypothesis using domain knowledge acquired from the training set through a similarity analysis, exploring the projected feature space to examine the class centroid shifts. The methodology is tested on a binary classification problem, where synthetic images of cubes and cylinders in different orientations are generated. The methodology improves the accuracy of the object classifier from 60% to around 90% in the case of a domain shift in physical feature space without human labeling. Full article

Generative Adversarial Networks are powerful generative models that are used in different areas and with multiple applications. However, this type of model has a training problem called mode collapse. This problem causes the generator to not learn the complete distribution of the data with which it is trained. To force the network to learn the entire data distribution, MSSGAN is introduced. This model has multiple generators and distributes the training data in multiple subspaces, where each generator is enforced to learn only one of the groups with the help of a classifier. We demonstrate that our model performs better on the FID and Sample Distribution metrics compared to previous models to avoid mode collapse. Experimental results show how each of the generators learns different information and, in turn, generates satisfactory quality samples. Full article

Reinforcement Learning (RL) has shown promise in optimizing complex control and decision-making processes but Deep Reinforcement Learning (DRL) lacks interpretability, limiting its adoption in regulated sectors like manufacturing, finance, and healthcare. Difficulties arise from DRL’s opaque decision-making, hindering efficiency and resource use, this issue is amplified with every advancement. While many seek to move from Experience Replay to A3C, the latter demands more resources. Despite efforts to improve Experience Replay selection strategies, there is a tendency to keep the capacity high. We investigate training a Deep Convolutional Q-learning agent across 20 Atari games intentionally reducing Experience Replay capacity from $1\times {10}^6$ to $5\times {10}^2$. We find that a reduction from $1\times {10}^4$ to $5\times {10}^3$ doesn’t significantly affect rewards, offering a practical path to resource-efficient DRL. To illuminate agent decisions and align them with game mechanics, we employ a novel method: visualizing Experience Replay via Deep SHAP Explainer. This approach fosters comprehension and transparent, interpretable explanations, though any capacity reduction must be cautious to avoid overfitting. Our study demonstrates the feasibility of reducing Experience Replay and advocates for transparent, interpretable decision explanations using the Deep SHAP Explainer to promote enhancing resource efficiency in Experience Replay. Full article

Analysis of data is crucial in waste management to improve effective planning from both short- and long-term perspectives. Real-world data often presents anomalies, but in the waste management sector, anomaly detection is seldom performed. The main goal and contribution of this paper is a proposal of a complex machine learning framework for changepoint detection in a large number of short time series from waste management. In such a case, it is not possible to use only an expert-based approach due to the time-consuming nature of this process and subjectivity. The proposed framework consists of two steps: (1) outlier detection via outlier test for trend-adjusted data, and (2) changepoints are identified via comparison of linear model parameters. In order to use the proposed method, it is necessary to have a sufficient number of experts’ assessments of the presence of anomalies in time series. The proposed framework is demonstrated on waste management data from the Czech Republic. It is observed that certain waste categories in specific regions frequently exhibit changepoints. On the micro-regional level, approximately 31.1% of time series contain at least one outlier and 16.4% exhibit changepoints. Certain groups of waste are more prone to the occurrence of anomalies. The results indicate that even in the case of aggregated data, anomalies are not rare, and their presence should always be checked. Full article

Federated learning (FL) is a technique that allows multiple participants to collaboratively train a Deep Neural Network (DNN) without the need to centralize their data. Among other advantages, it comes with privacy-preserving properties, making it attractive for application in sensitive contexts, such as health care or the military. Although the data are not explicitly exchanged, the training procedure requires sharing information about participants’ models. This makes the individual models vulnerable to theft or unauthorized distribution by malicious actors. To address the issue of ownership rights protection in the context of machine learning (ML), DNN watermarking methods have been developed during the last five years. Most existing works have focused on watermarking in a centralized manner, but only a few methods have been designed for FL and its unique constraints. In this paper, we provide an overview of recent advancements in federated learning watermarking, shedding light on the new challenges and opportunities that arise in this field. Full article

This paper addresses the problem of learning temporal graph representations, which capture the changing nature of complex evolving networks. Existing approaches mainly focus on adding new nodes and edges to capture dynamic graph structures. However, to achieve more accurate representation of graph evolution, we consider both the addition and deletion of nodes and edges as events. These events occur at irregular time scales and are modeled using temporal point processes. Our goal is to learn the conditional intensity function of the temporal point process to investigate the influence of deletion events on node representation learning for link-level prediction. We incorporate network entropy, a measure of node and edge significance, to capture the effect of node deletion and edge removal in our framework. Additionally, we leveraged the characteristics of a generalized temporal Hawkes process, which considers the inhibitory effects of events where past occurrences can reduce future intensity. This framework enables dynamic representation learning by effectively modeling both addition and deletion events in the temporal graph. To evaluate our approach, we utilize autonomous system graphs, a family of inhomogeneous sparse graphs with instances of node and edge additions and deletions, in a link prediction task. By integrating these enhancements into our framework, we improve the accuracy of dynamic link prediction and enable better understanding of the dynamic evolution of complex networks. Full article

Long-term dependence is an essential feature for the predictability of time series. Estimating the parameter that describes long memory is essential to describing the behavior of time series models. However, most long memory estimation methods assume that this parameter has a constant value throughout the time series, and do not consider that the parameter may change over time. In this work, we propose an automated methodology that combines the estimation methodologies of the fractional differentiation parameter (and/or Hurst parameter) with its application to Recurrent Neural Networks (RNNs) in order for said networks to learn and predict long memory dependencies from information obtained in nonlinear time series. The proposal combines three methods that allow for better approximation in the prediction of the values of the parameters for each one of the windows obtained, using Recurrent Neural Networks as an adaptive method to learn and predict the dependencies of long memory in Time Series. For the RNNs, we have evaluated four different architectures: the Simple RNN, LSTM, the BiLSTM, and the GRU. These models are built from blocks with gates controlling the cell state and memory. We have evaluated the proposed approach using both synthetic and real-world data sets. We have simulated ARFIMA models for the synthetic data to generate several time series by varying the fractional differentiation parameter. We have evaluated the proposed approach using synthetic and real datasets using Whittle’s estimates of the Hurst parameter classically obtained in each window. We have simulated ARFIMA models in such a way that the synthetic data generate several time series by varying the fractional differentiation parameter. The real-world IPSA stock option index and Tree Ringtime series datasets were evaluated. All of the results show that the proposed approach can predict the Hurst exponent with good performance by selecting the optimal window size and overlap change. Full article

This study introduces an optimal topology of vision transformers for real-time video action recognition in a cloud-based solution. Although model performance is a key criterion for real-time video analysis use cases, inference latency plays a more crucial role in adopting such technology in real-world scenarios. Our objective is to reduce the inference latency of the solution while admissibly maintaining the vision transformer’s performance. Thus, we employed the optimal cloud components as the foundation of our machine learning pipeline and optimized the topology of vision transformers. We utilized UCF101, including more than one million action recognition video clips. The modeling pipeline consists of a preprocessing module to extract frames from video clips, training two-dimensional (2D) vision transformer models, and deep learning baselines. The pipeline also includes a postprocessing step to aggregate the frame-level predictions to generate the video-level predictions at inference. The results demonstrate that our optimal vision transformer model with an input dimension of 56 × 56 × 3 with eight attention heads produces an F1 score of 91.497% for the testing set. The optimized vision transformer reduces the inference latency by 40.70%, measured through a batch-processing approach, with a 55.63% faster training time than the baseline. Lastly, we developed an enhanced skip-frame approach to improve the inference latency by finding an optimal ratio of frames for prediction at inference, where we could further reduce the inference latency by 57.15%. This study reveals that the vision transformer model is highly optimizable for inference latency while maintaining the model performance. Full article

A critical unmet medical need in prostate cancer (PCa) clinical management centers around distinguishing indolent from aggressive tumors. Traditionally, Gleason grading has been utilized for this purpose. However, tumor classification using Gleason Grade 7 is often ambiguous, as the clinical behavior of these tumors follows a variable clinical course. This study aimed to investigate the application of machine learning techniques (ML) to classify patients into indolent and aggressive PCas. We used gene expression data from The Cancer Genome Atlas and compared gene expression levels between indolent and aggressive tumors to identify features for developing and validating a range of ML and stacking algorithms. ML algorithms accurately distinguished indolent from aggressive PCas. With the accuracy of 96%, the stacking model was superior to individual ML algorithms when all samples with primary Gleason Grades 6 to 10 were used. Excluding samples with Gleason Grade 7 improved accuracy to 97%. This study shows that ML algorithms and stacking models are powerful approaches for the accurate classification of indolent versus aggressive PCas. Future implementation of this methodology may significantly impact clinical decision making and patient outcomes in the clinical management of prostate cancer. Full article

A modification of the brainstorming process by the application of artificial intelligence (AI) was proposed. Here, we describe the design of the software system “kresilnik”, which enables hybrid work between a human group and AI. The proposed system integrates the Open AI-GPT-3.5–turbo model with the server side providing the results to clients. The proposed architecture provides the possibility to not only generate ideas but also categorize them and set priorities. With the developed prototype, 760 ideas were generated on the topic of the design of the Gorenjska region’s development plan with eight different temperatures with the OpenAI-GPT-3.5-turbo algorithm. For the set of generated ideas, the entropy was determined, as well as the time needed for their generation. The distributions of the entropy of the ideas generated by the human-generated and the AI-generated sets of ideas of the OpenAI-GPT-3.5–turbo algorithm at different temperatures are provided in the form of histograms. Ideas are presented as word clouds and histograms for the human group and the AI-generated sets. A comparison of the process of generating ideas between the human group and AI was conducted. The statistical Mann-Whitney U-test was performed, which confirmed the significant differences in the average entropy of the generated ideas. Correlations between the length of the generated ideas and the time needed were determined for the human group and AI. The distributions for the time needed and the length of the ideas were determined, which are possible indicators to distinguish between human and artificial processes of generating ideas. Full article

Machine learning (ML) has been used in different ways in the fight against COVID-19 disease. ML models have been developed, e.g., for diagnostic or prognostic purposes and using various modalities of data (e.g., textual, visual, or structured). Due to the many specific aspects of this disease and its evolution over time, there is still not enough understanding of all relevant factors influencing the course of COVID-19 in particular patients. In all aspects of our work, there was a strong involvement of a medical expert following the human-in-the-loop principle. This is a very important but usually neglected part of the ML and knowledge extraction (KE) process. Our research shows that explainable artificial intelligence (XAI) may significantly support this part of ML and KE. Our research focused on using ML for knowledge extraction in two specific scenarios. In the first scenario, we aimed to discover whether adding information about the predominant COVID-19 variant impacts the performance of the ML models. In the second scenario, we focused on prognostic classification models concerning the need for an intensive care unit for a given patient in connection with different explainability AI (XAI) methods. We have used nine ML algorithms, namely XGBoost, CatBoost, LightGBM, logistic regression, Naive Bayes, random forest, SGD, SVM-linear, and SVM-RBF. We measured the performance of the resulting models using precision, accuracy, and AUC metrics. Subsequently, we focused on knowledge extraction from the best-performing models using two different approaches as follows: (a) features extracted automatically by forward stepwise selection (FSS); (b) attributes and their interactions discovered by model explainability methods. Both were compared with the attributes selected by the medical experts in advance based on the domain expertise. Our experiments showed that adding information about the COVID-19 variant did not influence the performance of the resulting ML models. It also turned out that medical experts were much more precise in the identification of significant attributes than FSS. Explainability methods identified almost the same attributes as a medical expert and interesting interactions among them, which the expert discussed from a medical point of view. The results of our research and their consequences are discussed. Full article

Identifying similar network structures is key to capturing graph isomorphisms and learning representations that exploit structural information encoded in graph data. This work shows that ego networks can produce a structural encoding scheme for arbitrary graphs with greater expressivity than the Weisfeiler–Lehman (1-WL) test. We introduce $\mathrm{I}\mathrm{G}\mathrm{E}\mathrm{L}$, a preprocessing step to produce features that augment node representations by encoding ego networks into sparse vectors that enrich message passing (MP) graph neural networks (GNNs) beyond 1-WL expressivity. We formally describe the relation between $\mathrm{I}\mathrm{G}\mathrm{E}\mathrm{L}$ and 1-WL, and characterize its expressive power and limitations. Experiments show that $\mathrm{I}\mathrm{G}\mathrm{E}\mathrm{L}$ matches the empirical expressivity of state-of-the-art methods on isomorphism detection while improving performance on nine GNN architectures and six graph machine learning tasks. Full article