Multiclass feature selection with metaheuristic optimization algorithms: a review

Greater cane rat algorithm (GCRA): A nature-inspired metaheuristic for optimization problems

This paper introduces a new metaheuristic technique known as the Greater Cane Rat Algorithm (GCRA) for addressing optimization problems. The optimization process of GCRA is inspired by the intelligent foraging behaviors of greater cane rats during and off mating season. Being highly nocturnal, they are intelligible enough to leave trails as they forage through reeds and grass. Such trails would subsequently lead to food and water sources and shelter. The exploration phase is achieved when they leave the different shelters scattered around their territory to forage and leave trails. It is presumed that the alpha male maintains knowledge about these routes, and as a result, other rats modify their location according to this information. Also, the males are aware of the breeding season and separate themselves from the group. The assumption is that once the group is separated during this season, the foraging activities are concentrated within areas of abundant food sources, which aids the exploitation. Hence, the smart foraging paths and behaviors during the mating season are mathematically represented to realize the design of the GCR algorithm and carry out the optimization tasks. The performance of GCRA is tested using twenty-two classical benchmark functions, ten CEC 2020 complex functions, and the CEC 2011 real-world continuous benchmark problems. To further test the performance of the proposed algorithm, six classic problems in the engineering domain were used. Furthermore, a thorough analysis of computational and convergence results is presented to shed light on the efficacy and stability levels of GCRA. The statistical significance of the results is compared with ten state-of-the-art algorithms using Friedman's and Wilcoxon's signed rank tests. These findings show that GCRA produced optimal or nearly optimal solutions and evaded the trap of local minima, distinguishing it from the rival optimization algorithms employed to tackle similar problems. The GCRA optimizer source code is publicly available at: https://www.mathworks.com/matlabcentral/fileexchange/165241-greater-cane-rat-algorithm-gcra.

Algorithm for community security risk assessment and influencing factors analysis by back propagation neural network

This paper aims to accurately assess and effectively manage various security risks in the community and overcome the challenges faced by traditional models in handling large amounts of features and high-dimensional data. Hence, this paper utilizes the back propagation neural network (BPNN) to optimize the security risk assessment model. A key challenge of researching community security risk assessment lies in accurately identifying and predicting a range of potential security threats. These threats may encompass natural disasters, public health crises, accidents, and social security issues. The intricate interplay of these risk factors, combined with the dynamic nature of community environments, presents difficulties for traditional risk assessment methodologies to address effectively. Initially, this paper delves into the factors influencing safety incidents within communities and establishes a comprehensive system of safety risk assessment indicators. Leveraging the adaptable and generalizable nature of the BPNN model, the paper proceeds to optimize the BPNN model, enhancing the security risk assessment model through this optimization. Subsequent comparison experiments with traditional models validate the rationality and effectiveness of the proposed model, with hidden layer nodes set at various levels like 10, 15, 20, 25, 30, and 35. These traditional models include Convolutional Neural Network (CNN), Long Short-Term Memory Network (LSTM), Bidirectional Encoder Representations from Transformers (BERT), Generative Pre-trained Transformer (GPT), and eXtreme Gradient Boosting (XGBOOST). Experimental findings demonstrate that with 20 hidden layer nodes, the optimized model achieves a remarkable final recognition accuracy of 99.1 %. Moreover, the optimized model exhibits significantly lower final function loss compared to models with different node numbers. Increasing the number of hidden layer nodes may diminish the optimized model's fit and accuracy. Comparison with traditional models reveals that the average accuracy of the optimized model in community risk identification reaches 98.5 %, with a maximum accuracy of 99.6 %. This marks an improvement of 9%-11 % in recognition accuracy across various risk factors compared to traditional models. Regarding system response time and resource utilization, the optimized model exhibits a response time ranging from 100 ms to 120 ms and consistently lower resource utilization rates across all scenarios, underscoring its efficiency in community security risk assessment. In conclusion, this experiment sheds light on the underlying mechanisms and patterns of community safety risk formation, offering novel perspectives and methodologies for researching community safety risk assessment. The paper concludes by presenting recommendations and strategies for addressing community safety risks based on experimental analysis.

Proposal and Definition of an Intelligent Clinical Decision Support System Applied to the Prediction of Dyspnea after 12 Months of an Acute Episode of COVID-19

Long COVID is a condition that affects a significant proportion of patients who have had COVID-19. It is characterised by the persistence of associated symptoms after the acute phase of the illness has subsided. Although several studies have investigated the risk factors associated with long COVID, identifying which patients will experience long-term symptoms remains a complex task. Among the various symptoms, dyspnea is one of the most prominent due to its close association with the respiratory nature of COVID-19 and its disabling consequences. This work proposes a new intelligent clinical decision support system to predict dyspnea 12 months after a severe episode of COVID-19 based on the SeguiCovid database from the Álvaro Cunqueiro Hospital in Vigo (Galicia, Spain). The database is initially processed using a CART-type decision tree to identify the variables with the highest predictive power. Based on these variables, a cascade of expert systems has been defined with Mamdani-type fuzzy-inference engines. The rules for each system were generated using the Wang-Mendel automatic rule generation algorithm. At the output of the cascade, a risk indicator is obtained, which allows for the categorisation of patients into two groups: those with dyspnea and those without dyspnea at 12 months. This simplifies follow-up and the performance of studies aimed at those patients at risk. The system has produced satisfactory results in initial tests, supported by an AUC of 0.75, demonstrating the potential and usefulness of this tool in clinical practice.

S-shaped grey wolf optimizer-based FOX algorithm for feature selection

The FOX algorithm is a recently developed metaheuristic approach inspired by the behavior of foxes in their natural habitat. While the FOX algorithm exhibits commendable performance, its basic version, in complex problem scenarios, may become trapped in local optima, failing to identify the optimal solution due to its weak exploitation capabilities. This research addresses a high-dimensional feature selection problem. In feature selection, the most informative features are retained while discarding irrelevant ones. An enhanced version of the FOX algorithm is proposed, aiming to mitigate its drawbacks in feature selection. The improved approach referred to as S-shaped Grey Wolf Optimizer-based FOX (FOX-GWO), which focuses on augmenting the local search capabilities of the FOX algorithm via the integration of GWO. Additionally, the introduction of an S-shaped transfer function enables the population to explore both binary options throughout the search process. Through a series of experiments on 18 datasets with varying dimensions, FOX-GWO outperforms in 83.33 % of datasets for average accuracy, 61.11 % for reduced feature dimensionality, and 72.22 % for average fitness value across the 18 datasets. Meaning it efficiently explores high-dimensional spaces. These findings highlight its practical value and potential to advance feature selection in complex data analysis, enhancing model prediction accuracy.

Feature Selection Problem and Metaheuristics: A Systematic Literature Review about Its Formulation, Evaluation and Applications

Feature selection is becoming a relevant problem within the field of machine learning. The feature selection problem focuses on the selection of the small, necessary, and sufficient subset of features that represent the general set of features, eliminating redundant and irrelevant information. Given the importance of the topic, in recent years there has been a boom in the study of the problem, generating a large number of related investigations. Given this, this work analyzes 161 articles published between 2019 and 2023 (20 April 2023), emphasizing the formulation of the problem and performance measures, and proposing classifications for the objective functions and evaluation metrics. Furthermore, an in-depth description and analysis of metaheuristics, benchmark datasets, and practical real-world applications are presented. Finally, in light of recent advances, this review paper provides future research opportunities.

Clinical-functional brain connectivity signature predicts longitudinal symptom improvement after acupuncture treatment in patients with functional dyspepsia

Whilst acupuncture has been shown to be an effective treatment for functional dyspepsia (FD), its efficacy varies significantly among patients. Knowing beforehand how each patient responds to acupuncture treatment will facilitate the ability to produce personalized prescriptions, therefore, improving acupuncture efficacy. The objective of this study was to construct the prediction model, based on the clinical-neuroimaging signature, to forecast the individual symptom improvement of FD patients following a 4-week acupuncture treatment and to identify the critical predictive features that could potentially serve as biomarkers for predicting the efficacy of acupuncture for FD. Clinical-functional brain connectivity signatures were extracted from samples in the training-test set (100 FD patients) and independent validation set (60 FD patients). Based on these signatures and support vector machine algorithms, prediction models were developed in the training test set, followed by model performance evaluation and predictive features extraction. Subsequently, the external robustness of the extracted predictive features in predicting acupuncture efficacy was evaluated by the independent validation set. The developed prediction models possessed an accuracy of 88% in predicting acupuncture responders, as well as an R2 of 0.453 in forecasting symptom relief. Factors that contributed significantly to stronger responsiveness of patients to acupuncture therapy included higher resting-state functional connectivity associated with the orbitofrontal gyrus, caudate, hippocampus, and anterior insula, as well as higher baseline scores of the Symptom Index of Dyspepsia and shorter durations of the condition. Furthermore, the robustness of these features in predicting the efficacy of acupuncture for FD was verified through various machine learning algorithms and independent samples and remained stable in univariate and multivariate analyses. These findings suggest that it is both feasible and reliable to predict the efficacy of acupuncture for FD based on the pre-treatment clinical-neuroimaging signature. The established prediction framework will promote the identification of suitable candidates for acupuncture treatment, thereby improving the efficacy and reducing the cost of acupuncture for FD.

Understanding the Role of Sensor Optimisation in Complex Systems

Complex systems involve monitoring, assessing, and predicting the health of various systems within an integrated vehicle health management (IVHM) system or a larger system. Health management applications rely on sensors that generate useful information about the health condition of the assets; thus, optimising the sensor network quality while considering specific constraints is the first step in assessing the condition of assets. The optimisation problem in sensor networks involves considering trade-offs between different performance metrics. This review paper provides a comprehensive guideline for practitioners in the field of sensor optimisation for complex systems. It introduces versatile multi-perspective cost functions for different aspects of sensor optimisation, including selection, placement, data processing and operation. A taxonomy and concept map of the field are defined as valuable navigation tools in this vast field. Optimisation techniques and quantification approaches of the cost functions are discussed, emphasising their adaptability to tailor to specific application requirements. As a pioneering contribution, all the relevant literature is gathered and classified here to further improve the understanding of optimal sensor networks from an information-gain perspective.

Utilizing convolutional neural networks to classify monkeypox skin lesions

Monkeypox is a rare viral disease that can cause severe illness in humans, presenting with skin lesions and rashes. However, accurately diagnosing monkeypox based on visual inspection of the lesions can be challenging and time-consuming, especially in resource-limited settings where laboratory tests may not be available. In recent years, deep learning methods, particularly Convolutional Neural Networks (CNNs), have shown great potential in image recognition and classification tasks. To this end, this study proposes an approach using CNNs to classify monkeypox skin lesions. Additionally, the study optimized the CNN model using the Grey Wolf Optimizer (GWO) algorithm, resulting in a significant improvement in accuracy, precision, recall, F1-score, and AUC compared to the non-optimized model. The GWO optimization strategy can enhance the performance of CNN models on similar tasks. The optimized model achieved an impressive accuracy of 95.3%, indicating that the GWO optimizer has improved the model's ability to discriminate between positive and negative classes. The proposed approach has several potential benefits for improving the accuracy and efficiency of monkeypox diagnosis and surveillance. It could enable faster and more accurate diagnosis of monkeypox skin lesions, leading to earlier detection and better patient outcomes. Furthermore, the approach could have crucial public health implications for controlling and preventing monkeypox outbreaks. Overall, this study offers a novel and highly effective approach for diagnosing monkeypox, which could have significant real-world applications.

A New Fuzzy-Based Classification Method for Use in Smart/Precision Medicine

The development of information technology has had a significant impact on various areas of human activity, including medicine. It has led to the emergence of the phenomenon of Industry 4.0, which, in turn, led to the development of the concept of Medicine 4.0. Medicine 4.0, or smart medicine, can be considered as a structural association of such areas as AI-based medicine, telemedicine, and precision medicine. Each of these areas has its own characteristic data, along with the specifics of their processing and analysis. Nevertheless, at present, all these types of data must be processed simultaneously, in order to provide the most complete picture of the health of each individual patient. In this paper, after a brief analysis of the topic of medical data, a new classification method is proposed that allows the processing of the maximum number of data types. The specificity of this method is its use of a fuzzy classifier. The effectiveness of this method is confirmed by an analysis of the results from the classification of various types of data for medical applications and health problems. In this paper, as an illustration of the proposed method, a fuzzy decision tree has been used as the fuzzy classifier. The accuracy of the classification in terms of the proposed method, based on a fuzzy classifier, gives the best performance in comparison with crisp classifiers.

Improved Reptile Search Algorithm by Salp Swarm Algorithm for Medical Image Segmentation

This study proposes a novel nature-inspired meta-heuristic optimizer based on the Reptile Search Algorithm combed with Salp Swarm Algorithm for image segmentation using gray-scale multi-level thresholding, called RSA-SSA. The proposed method introduces a better search space to find the optimal solution at each iteration. However, we proposed RSA-SSA to avoid the searching problem in the same area and determine the optimal multi-level thresholds. The obtained solutions by the proposed method are represented using the image histogram. The proposed RSA-SSA employed Otsu's variance class function to get the best threshold values at each level. The performance measure for the proposed method is valid by detecting fitness function, structural similarity index, peak signal-to-noise ratio, and Friedman ranking test. Several benchmark images of COVID-19 validate the performance of the proposed RSA-SSA. The results showed that the proposed RSA-SSA outperformed other metaheuristics optimization algorithms published in the literature.

Interpretable machine learning with tree-based shapley additive explanations: Application to metabolomics datasets for binary classification

Machine learning (ML) models are used in clinical metabolomics studies most notably for biomarker discoveries, to identify metabolites that discriminate between a case and control group. To improve understanding of the underlying biomedical problem and to bolster confidence in these discoveries, model interpretability is germane. In metabolomics, partial least square discriminant analysis (PLS-DA) and its variants are widely used, partly due to the model's interpretability with the Variable Influence in Projection (VIP) scores, a global interpretable method. Herein, Tree-based Shapley Additive explanations (SHAP), an interpretable ML method grounded in game theory, was used to explain ML models with local explanation properties. In this study, ML experiments (binary classification) were conducted for three published metabolomics datasets using PLS-DA, random forests, gradient boosting, and extreme gradient boosting (XGBoost). Using one of the datasets, PLS-DA model was explained using VIP scores, while one of the best-performing models, a random forest model, was interpreted using Tree SHAP. The results show that SHAP has a more explanation depth than PLS-DA's VIP, making it a powerful method for rationalizing machine learning predictions from metabolomics studies.

Improved Dwarf Mongoose Optimization for Constrained Engineering Design Problems

This paper proposes a modified version of the Dwarf Mongoose Optimization Algorithm (IDMO) for constrained engineering design problems. This optimization technique modifies the base algorithm (DMO) in three simple but effective ways. First, the alpha selection in IDMO differs from the DMO, where evaluating the probability value of each fitness is just a computational overhead and contributes nothing to the quality of the alpha or other group members. The fittest dwarf mongoose is selected as the alpha, and a new operator ω is introduced, which controls the alpha movement, thereby enhancing the exploration ability and exploitability of the IDMO. Second, the scout group movements are modified by randomization to introduce diversity in the search process and explore unvisited areas. Finally, the babysitter's exchange criterium is modified such that once the criterium is met, the babysitters that are exchanged interact with the dwarf mongoose exchanging them to gain information about food sources and sleeping mounds, which could result in better-fitted mongooses instead of initializing them afresh as done in DMO, then the counter is reset to zero. The proposed IDMO was used to solve the classical and CEC 2020 benchmark functions and 12 continuous/discrete engineering optimization problems. The performance of the IDMO, using different performance metrics and statistical analysis, is compared with the DMO and eight other existing algorithms. In most cases, the results show that solutions achieved by the IDMO are better than those obtained by the existing algorithms.

An Al-Biruni Earth Radius Optimization-Based Deep Convolutional Neural Network for Classifying Monkeypox Disease

Human skin diseases have become increasingly prevalent in recent decades, with millions of individuals in developed countries experiencing monkeypox. Such conditions often carry less obvious but no less devastating risks, including increased vulnerability to monkeypox, cancer, and low self-esteem. Due to the low visual resolution of monkeypox disease images, medical specialists with high-level tools are typically required for a proper diagnosis. The manual diagnosis of monkeypox disease is subjective, time-consuming, and labor-intensive. Therefore, it is necessary to create a computer-aided approach for the automated diagnosis of monkeypox disease. Most research articles on monkeypox disease relied on convolutional neural networks (CNNs) and using classical loss functions, allowing them to pick up discriminative elements in monkeypox images. To enhance this, a novel framework using Al-Biruni Earth radius (BER) optimization-based stochastic fractal search (BERSFS) is proposed to fine-tune the deep CNN layers for classifying monkeypox disease from images. As a first step in the proposed approach, we use deep CNN-based models to learn the embedding of input images in Euclidean space. In the second step, we use an optimized classification model based on the triplet loss function to calculate the distance between pairs of images in Euclidean space and learn features that may be used to distinguish between different cases, including monkeypox cases. The proposed approach uses images of human skin diseases obtained from an African hospital. The experimental results of the study demonstrate the proposed framework's efficacy, as it outperforms numerous examples of prior research on skin disease problems. On the other hand, statistical experiments with Wilcoxon and analysis of variance (ANOVA) tests are conducted to evaluate the proposed approach in terms of effectiveness and stability. The recorded results confirm the superiority of the proposed method when compared with other optimization algorithms and machine learning models.

Immunity-based Ebola optimization search algorithm for minimization of feature extraction with reduction in digital mammography using CNN models

Feature classification in digital medical images like mammography presents an optimization problem which researchers often neglect. The use of a convolutional neural network (CNN) in feature extraction and classification has been widely reported in the literature to have achieved outstanding performance and acceptance in the disease detection procedure. However, little emphasis is placed on ensuring that only discriminant features extracted by the convolutional operations are passed on to the classifier, to avoid bottlenecking the classification operation. Unfortunately, since this has been left unaddressed, a subtle performance impairment has resulted from this omission. Therefore, this study is devoted to addressing these drawbacks using a metaheuristic algorithm to optimize the number of features extracted by the CNN, so that suggestive features are applied for the classification process. To achieve this, a new variant of the Ebola-based optimization algorithm is proposed, based on the population immunity concept and the use of a chaos mapping initialization strategy. The resulting algorithm, called the immunity-based Ebola optimization search algorithm (IEOSA), is applied to the optimization problem addressed in the study. The optimized features represent the output from the IEOSA, which receives the noisy and unfiltered detected features from the convolutional process as input. An exhaustive evaluation of the IEOSA was carried out using classical and IEEE CEC benchmarked functions. A comparative analysis of the performance of IEOSA is presented, with some recent optimization algorithms. The experimental result showed that IEOSA performed well on all the tested benchmark functions. Furthermore, IEOSA was then applied to solve the feature enhancement and selection problem in CNN for better prediction of breast cancer in digital mammography. The classification accuracy returned by the IEOSA method showed that the new approach improved the classification process on detected features when using CNN models.

Stability analysis with general fuzzy measure: An application to social security organizations

An effective method for evaluating the efficiency of peer decision-making units (DMUs) is data envelope analysis (DEA). In engineering sciences and real-world management problems, uncertainty in input and output data always exists. To achieve reliable results, uncertainties must be taken into account. In this research, a General Fuzzy (GF) approach is designed to cope with uncertainty in the presence of fuzzy observations for categorizing and specifying stability radius and alterations ranges of efficient and inefficient DMUs, which is applicable to real-world decision-making problems. For this purpose, a DEA sensitivity analysis model is presented, which will be modeled by fuzzy sets. Then, by applying the General Fuzzy (GF) approach, the fuzzy DEA sensitivity analysis model is transformed into the equivalent crisp form of fuzzy chance constraints according to specific confidence levels. Finally, a numerical example and a case study of branches of the social security organization are presented to illustrate sensitivity and stability analysis in the presence of fuzzy data. The obtained results provide the input and output changes of the evaluated units according to the attitude and preference of the decision maker with different confidence levels so that the data changes in the fuzzy environment do not change the units' classification from efficient to inefficient and vice versa.

Binary dwarf mongoose optimizer for solving high-dimensional feature selection problems

Selecting appropriate feature subsets is a vital task in machine learning. Its main goal is to remove noisy, irrelevant, and redundant feature subsets that could negatively impact the learning model's accuracy and improve classification performance without information loss. Therefore, more advanced optimization methods have been employed to locate the optimal subset of features. This paper presents a binary version of the dwarf mongoose optimization called the BDMO algorithm to solve the high-dimensional feature selection problem. The effectiveness of this approach was validated using 18 high-dimensional datasets from the Arizona State University feature selection repository and compared the efficacy of the BDMO with other well-known feature selection techniques in the literature. The results show that the BDMO outperforms other methods producing the least average fitness value in 14 out of 18 datasets which means that it achieved 77.77% on the overall best fitness values. The result also shows BDMO demonstrating stability by returning the least standard deviation (SD) value in 13 of 18 datasets (72.22%). Furthermore, the study achieved higher validation accuracy in 15 of the 18 datasets (83.33%) over other methods. The proposed approach also yielded the highest validation accuracy attainable in the COIL20 and Leukemia datasets which vividly portray the superiority of the BDMO.

A Comprehensive Review of Bat Inspired Algorithm: Variants, Applications, and Hybridization

Bat algorithm (BA) is one of the promising metaheuristic algorithms. It proved its efficiency in dealing with various optimization problems in diverse fields, such as power and energy systems, economic load dispatch problems, engineering design, image processing and medical applications. Thus, this review introduces a comprehensive and exhaustive review of the BA, as well as evaluates its main characteristics by comparing it with other optimization algorithms. The review paper highlights the performance of BA in different applications and the modifications that have been conducted by researchers (i.e., variants of BA). At the end, the conclusions focus on the current work on BA, highlighting its weaknesses, and suggest possible future research directions. The review paper will be helpful for the researchers and practitioners of BA belonging to a wide range of audiences from the domains of optimization, engineering, medical, data mining and clustering. As well, it is wealthy in research on health, environment and public safety. Also, it will aid those who are interested by providing them with potential future research.