Search

A main research direction in the field of evolutionary machine learning is to develop a scalable classifier system to solve high-dimensional problems. Recently work has begun on autonomously reusing learned building blocks of knowledge to scale from low-dimensional problems to high-dimensional ones. An XCS-based classifier system, known as XCSCFC, has been shown to be scalable, through the addition of expression tree-like code fragments, to a limit beyond standard learning classifier systems. XCSCFC is especially beneficial if the target problem can be divided into a hierarchy of subproblems and each of them is solvable in a bottom-up fashion. However, if the hierarchy of subproblems is too deep, then XCSCFC becomes impractical because of the needed computational time and thus eventually hits a limit in problem size. A limitation in this technique is the lack of a cyclic representation, which is inherent in finite state machines (FSMs). However, the evolution of FSMs is a hard task owing to the combinatorially large number of possible states, connections, and interaction. Usually this requires supervised learning to minimize inappropriate FSMs, which for high-dimensional problems necessitates subsampling or incremental testing. To avoid these constraints, this work introduces a state-machine-based encoding scheme into XCS for the first time, termed XCSSMA. The proposed system has been tested on six complex Boolean problem domains: multiplexer, majority-on, carry, even-parity, count ones, and digital design verification problems. The proposed approach outperforms XCSCFA (an XCS that computes actions) and XCSF (an XCS that computes predictions) in three of the six problem domains, while the performance in others is similar. In addition, XCSSMA evolved, for the first time, compact and human readable general classifiers (i.e., solving any n-bit problems) for the even-parity and carry problem domains, demonstrating its ability to produce scalable solutions using a cyclic representation.

Machine learning based on genetic algorithms is an important application. The multi-dimensional ordered sample clustering problem is often solved using Fisher's optimal segmentation method. However, this method has obvious shortcomings when encountering long sample problems due to its high storage requirements during the computation process. Therefore, Fisher's optimal two-segmentation method is generally used in practical problems instead, which avoids storage problems. But it is prone to local optima. Based on the analysis of the shortcomings of the Fisher optimal segmentation and optimal two-segmentation algorithms, this paper proposes a genetic-based machine learning clustering algorithm, which overcomes the problem of Fisher's optimal two-segmentation algorithm being prone to local optima and also solves the problem of high storage requirements during the computation process of Fisher's optimal segmentation method. The application of this algorithm in the optimization system of water environment monitoring points shows that it is effective. [ABSTRACT FROM AUTHOR]

The document is an author index for Volume 23 (2024) of the International Journal of Information Technology & Decision Making. It lists various authors and their respective articles published in the journal, covering topics such as Big Data Analytics, Machine Learning, IoT, and Decision Making. The index provides a comprehensive overview of the research articles and authors featured in the journal, offering insights into the diverse range of topics covered in the field of information technology and decision making. [Extracted from the article]

Entrepreneurs and small businesses have increasingly used reward-based crowdfunding to raise capital for their creative projects, whose success is central to this industry. Thus, predicting the success of crowdfunding campaigns is a topic of great importance for entrepreneurs and platform managers. The literature that employs monolithic classifiers and static ensemble learning for crowdfunding success prediction are scarce. In contrast, the dynamic selection (DS) algorithm, which belongs to the ensemble learning category, deserves a particular remark since it has overcome traditional monolithic classifiers and static ensembles in many applications. This paper proposes a dynamic selection framework for reward crowdfunding prediction. DS algorithms select a competent subset of the classifier per query instance. This procedure is performed during the generalization, and the subset is composed of local experts, favoring an increase in accuracy. Fifteen machine learning models are evaluated using three metrics (accuracy, area under the ROC curve and F-score), and ensemble learning obtained better results than traditional classifiers. In particular, Meta-DES, which performs dynamic selection, obtains the best overall results among the evaluated models. Furthermore, since usually interpreting the output of ML models is considered to be very difficult due to their complex "black box" architecture, we also use Shapley additive explanations to interpret the perdition's outputs. Among variables evaluated in our models, the textual sentiment of the mass media, the number of pledges, and the target amount of the campaign deserve a highlight when predicting the campaign's success. The source-code and further details about the experimental analysis are available at https://github.com/las33/Crowdfunding. [ABSTRACT FROM AUTHOR]

This paper describes and evaluates the performance of a learning classifier system (lcs) inspired algorithm called Temporal Reinforcement And Classification Architecture (traca) on maze navigation tasks which contain hidden state. The evaluation of traca includes comparisons with other learning algorithms on selected difficult maze navigation tasks. Not all lcss are capable of learning all types of hidden-state mazes so traca is specifically compared against selected other lcs-based approaches that are most capable on these tasks, including xcsmh, AgentP (G), and AgentP (SA). Each algorithm is evaluated using a maze navigation task that has been identified as among the most difficult due to recurring aliased regions. The comparisons between algorithms include training time, test performance, and the size of the learned rule sets. The results indicate that each algorithm has its own advantages and drawbacks. For example, on the most difficult maze traca's average steps to the goal are 10.1 while AgentP (G) are 7.87; however, traca requires an average of only 354 training trials compared with 537 for AgentP (G). Following the maze tasks, traca is also tested on two variations in a truck driving task where it must learn to navigate four lanes of slower vehicles while avoiding collisions. The results show that traca can achieve a low number of collisions with relatively few trials (as low as 24 collisions over 5000 time steps after 10,000 training time steps) but may require multiple network construction attempts to achieve high performance. [ABSTRACT FROM AUTHOR]

The lungs play a vital role in supplying oxygen to every cell, filtering air to prevent harmful substances, and supporting defense mechanisms. However, they remain susceptible to the risk of diseases such as infections, inflammation, and cancer that affect the lungs. Meta-ensemble techniques are prominent methods used in machine learning to enhance the accuracy of classifier learning systems in making predictions. This work proposes a robust predictive model using a meta-ensemble method to identify high-risk individuals with lung cancer, thereby taking early action to prevent longterm problems benchmarked upon the Kaggle Machine Learning practitioners' Lung Cancer Dataset. Three machine learning ensemble modelsdRandom Forest, Adaptive Boosting (AdaBoost), and Gradient Boostingdwere used to develop the meta-ensemble models proposed in this paper, whereby the three ensemble models were adopted as base classifiers while one of them was adopted as the meta-classifier. In addition, two of the ensemble models were used as base classifiers, while the third was used as a meta-classifier to evaluate lung cancer risk prediction. Different graphs were evaluated to show that people with these features are liable to develop lung cancer. The proposed model has immensely improved prediction performance. The meta-ensemble models were simulated using the Python simulation environment, and the 5-fold cross-validation technique was used. The model validation was carried out using several known performance evaluation methodologies. The results of the experiments showed that gradient boosting achieved a maximum accuracy of 100%, an area under the curve (AUC), and a precision of 100%. The proposed model was compared with novel machine learning methods and popular state-of-the-art (SOTA) deep learning techniques. It was confirmed from the results that the model in this study had the best accuracy at lung cancer risk prediction. This study's results can be utilized to enhance the performance of actual patient risk prediction systems in the future. [ABSTRACT FROM AUTHOR]

Background: Mortality prediction in interstitial lung disease (ILD) poses a significant challenge to clinicians due to heterogeneity across disease subtypes. Currently, forced vital capacity (FVC) and Gender, Age, and Physiology (GAP) score are the two most utilized metrics in prognostication. Recently, a machine learning classifier system, Fibresolve, designed to identify a variety of computed tomography (CT) patterns associated with idiopathic pulmonary fibrosis (IPF), was demonstrated to have a significant association with mortality across multiple subtypes of ILD. The purpose of this follow-up study was to retrospectively validate these findings in a large, external cohort of patients with ILD. Methods: In this multi-center validation study, Fibresolve was applied to chest CT scans of patients with confirmed ILD that had available follow-up data. Fibresolve scores categorized by tertile were analyzed using Cox regression analysis adjusted for tobacco use and modified GAP (mGAP) score. Results: Of 643 patients included, 446 (69.3%) died over a median follow-up time of 144 [1-821] weeks. The median [range] mGAP score was 5 [3–7]. In multivariable analysis, Fibresolve score categorized by tertile was significantly associated with mortality (Tertile 2 HR 1.47, 95% CI 0.82–2.37, p = 0.11; Tertile 3 HR 3.12, 95% CI 1.98–4.90, p < 0.001). Subgroup analyses revealed significant associations amongst those with non-IPF ILDs (Tertile 2 HR 1.95, 95% CI 1.28–2.97, Tertile 3 HR 4.66, 95% CI 2.94–7.38) and severe disease, defined by a FVC ≤ 75% (Tertile 2 HR 2.29, 95% CI 1.43–3.67, Tertile 3 HR 4.80, 95% CI 2.93–7.86). Conclusions: Fibresolve is independently associated with mortality in ILD, particularly amongst patients with non-IPF ILDs and in those with severe disease. [ABSTRACT FROM AUTHOR]

This document discusses the concept of lifelike computing systems, which aim to replicate qualities associated with living systems in technological systems. The document explains that lifelike computing systems draw inspiration from the field of Artificial Life (ALife) and integrate insights from various research initiatives, such as cybernetics, self-aware computing, and bio-inspired computing. The special issue presented in the document includes four papers that explore the integration of lifelike properties into engineered systems, covering topics such as metaheuristic methods, nature-inspired algorithms, explainable AI, and artificial collective intelligence engineering. The document concludes by expressing gratitude to the authors, reviewers, and organizers involved in the special issue. [Extracted from the article]

In sociotechnical settings, human operators are increasingly assisted by decision support systems. By employing such systems, important properties of sociotechnical systems, such as self-adaptation and self-optimization, are expected to improve further. To be accepted by and engage efficiently with operators, decision support systems need to be able to provide explanations regarding the reasoning behind specific decisions. In this article, we propose the use of learning classifier systems (LCSs), a family of rule-based machine learning methods, to facilitate and highlight techniques to improve transparent decision-making. Furthermore, we present a novel approach to assessing application-specific explainability needs for the design of LCS models. For this, we propose an application-independent template of seven questions. We demonstrate the approach's use in an interview-based case study for a manufacturing scenario. We find that the answers received do yield useful insights for a well-designed LCS model and requirements for stakeholders to engage actively with an intelligent agent. [ABSTRACT FROM AUTHOR]

Handwriting contains a lot of information about the person writing it and is a sign of personality traits represented by neurological patterns in the brain. In other words, our brain and subconscious actually shape our character as a result of our habits. It is possible to get an idea about the mood of the individual by examining the handwriting. Joy, sadness, anger and anxiety are some of them. In this study, handwritings of people belonging to different professions and age groups were collected. Feature extraction methods was applied on these articles by applying word and line detection, slant, pressure, page layout and similar image processing methods. The obtained features formed the inputs of the dataset. Personality traits such as calm, optimistic, emotional, extrovert, which were estimated using graphology, were added to the dataset as outputs. Then, this dataset was applied to Random Forest (RF), Naive Bayes (NB), Decision Tree, Support Vector Machines (SVM), Logistic Regression, Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) algorithms. According to the performance metrics used, the Random Forest algorithm gave the most successful results in terms of accuracy, precision and f1-score metrics. For this algorithm, the accuracy, precision, recall and f1 score values were found to be 0.90, 0.91, 0.84 and 0.85, respectively. Furthermore, the results of the personality analysis were compared with the results of the personality test performed by the expert psychologist. As a result of this comparison, it was seen that there was a 73% match. [ABSTRACT FROM AUTHOR]

As a component of Wireless Sensor Network (WSN), Visual-WSN (VWSN) utilizes cameras to obtain relevant data including visual recordings and static images. Data from the camera is sent to energy efficient sink to extract key-information out of it. VWSN applications range from health care monitoring to military surveillance. In a network with VWSN, there are multiple challenges to move high volume data from a source location to a target and the key challenges include energy, memory and I/O resources. In this case, Mobile Sinks (MS) can be employed for data collection which not only collects information from particular chosen nodes called Cluster Head (CH), it also collects data from nearby nodes as well. The innovation of our work is to intelligently decide on a particular node as CH whose selection criteria would directly have an impact on QoS parameters of the system. However, making an appropriate choice during CH selection is a daunting task as the dynamic and mobile nature of MSs has to be taken into account. We propose Genetic Machine Learning based Fuzzy system for clustering which has the potential to simulate human cognitive behavior to observe, learn and understand things from manual perspective. Proposed architecture is designed based on Mamdani's fuzzy model. Following parameters are derived based on the model residual energy, node centrality, distance between the sink and current position, node centrality, node density, node history, and mobility of sink as input variables for decision making in CH selection. The inputs received have a direct impact on the Fuzzy logic rules mechanism which in turn affects the accuracy of VWSN. The proposed work creates a mechanism to learn the fuzzy rules using Genetic Algorithm (GA) and to optimize the fuzzy rules base in order to eliminate irrelevant and repetitive rules. Genetic algorithmbased machine learning optimizes the interpretability aspect of fuzzy system. Simulation results are obtained using MATLAB. The result shows that the classification accuracy increase along with minimizing fuzzy rules count and thus it can be inferred that the suggested methodology has a better protracted lifetime in contrast with Low Energy Adaptive Clustering Hierarchy (LEACH) and LEACHExpected Residual Energy (LEACH-ERE). [ABSTRACT FROM AUTHOR]

Intrusion Detection Systems (IDS) are regarded as an efficient security mechanism in the Internet of Things (IoT). IDSs in the IoT face critical challenges, such as a vast amount of data with many features, limited resources, detection of unknown attacks, and many false alarms. To overcome these issues, this paper proposes a novel framework, namely a Multi-Layer Multi-Classification System (MLMCS). The proposed framework consists of four steps: In the first step, the data collected from information sources are pre-processed. The second step divides the feature space into several subspaces to reduce the model's complexity. The third step proposes a multilayer base classifier called a Multi-Layer Classifier System (MLCS) for each feature subspace. Each MLCS is layered to transform a multi-class classification into several binary classifications so that a specific category of attacks has been detected in each layer based on the Group Method of Data Handling (GMDH) neural network. In the fourth step, the results obtained from MLCSs based on the Dempster-Shafer theory are combined to reduce the problems related to overfitting. The simulation results were gained using three datasets, i.e., NSL-KDD, UNSW-NB15, and TON-IoT. They indicated that the proposed method, compared with other methods, improved F1-score by 5.64%, 0.65%, and 0.11%, respectively. Furthermore, the false alarm rate was reduced by 2.66%, 2.52%, and 0.02%. Moreover, the impact of four well-known adversarial attacks on the dataset was investigated. It was observed that the MLMCS method, with an average 5.21% reduction, could perform well against these attacks. [ABSTRACT FROM AUTHOR]

Pancreatic cancer is right now the fourth largest cause of cancer-related deaths. Early diagnosis is one good solution for pancreatic cancer patients and reduces the mortality rate. Accurate and earlier diagnosis of the pancreatic tumor is a demanding task due to several factors such as delayed diagnosis and absence of early warning symptoms. The conventional distributed machine learning techniques such as SVM and logistic regression were not efficient to minimize the error rate and improve the classification of pancreatic cancer with higher accuracy. Therefore, a novel technique called Distributed Hybrid Elitism gene Quadratic discriminant Reinforced Learning Classifier System (DHEGQDRLCS) is developed in this paper. First, the number of data samples is collected from the repository dataset. This repository contains all the necessary files for the identification of prognostic biomarkers for pancreatic cancer. After the data collection, the separation of training and testing samples is performed for the accurate classification of pancreatic cancer samples. Then the training samples are considered and applied to Distributed Hybrid Elitism gene Quadratic discriminant Reinforced Learning Classifier System. The proposed hybrid classifier system uses the Kernel Quadratic Discriminant Function to analyze the training samples. After that, the Elitism gradient gene optimization is applied for classifying the samples into multiple classes such as non-cancerous pancreas, benign hepatobiliary disease i.e., pancreatic cancer, and Pancreatic ductal adenocarcinoma. Then the Reinforced Learning technique is applied to minimize the loss function based on target classification results and predicted classification results. Finally, the hybridized approach improves pancreatic cancer diagnosing accuracy. Experimental evaluation is carried out with pancreatic cancer dataset with Hadoop distributed system and different quantitative metrics such as Accuracy, balanced accuracy, F1-score, precision, recall, specificity, TN, TP, FN, FP, ROCAUC, PRCAUC, and PRCAPS. The performance analysis results indicate that the DHEGQDRLCS provides better diagnosing accuracy when compared to existing methods. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a debilitating neurological disorder that most commonly affects the elderly. Alzheimer's disease patients have substantial memory loss. The Memory loss is caused by atrophy in the hippocampus, amygdala, and other areas of the brain in Alzheimer's patients. Identification and categorization of Alzheimer's disease are considered challenging research subjects due to the vast number of Alzheimer's patients and the absence of effective diagnostic procedures. And also traditional identification of Alzheimer's disease is taking more time consuming. In order to tackle this issue, we have to use the Artificial Intelligence technologies, which are driven by machine learning base AD classification and identification. The combined SIFT and SURF feature is used in this study to describe an AD Classification based on images. These combined feature parameters are then fed into a machine learning classifier for additional classification accuracy. There were three distinct machine learning classifiers compared to the system: SVM, DT and k-NN. For testing the proposed AD system, we have gathered a benchmark dataset that includes four categories: Mild, Moderate, Non Demented and Very Mild. The experimental results showed that SVM achieved higher accuracy than other classifiers. [ABSTRACT FROM AUTHOR]

Peroxidase mimicking Fe3O4@Chitosan (Fe3O4@Chi) nanozyme was synthesized and used for high-sensitive enzyme-free colorimetric detection of H2O2. The nanozyme was characterized in comparison with Fe3O4 nanoparticles (NPs) using X-ray diffraction, Fourier-transform infrared spectroscopy, dynamic light scattering, and thermogravimetric analysis. The catalytic performance of Fe3O4@Chi nanozyme was first evaluated by UV–Vis spectroscopy using 3,3′,5,5′-tetramethylbenzidine. Unlike Fe3O4NPs, Fe3O4@Chi nanozyme exhibited an intrinsic peroxidase activity with a detection limit of 69 nM. Next, the nanozyme was applied to a microfluidic paper-based analytical device (µPAD) and colorimetric analysis was performed at varying concentrations of H2O2 using a machine learning-based smartphone app called "Hi-perox Sens++." The app with machine learning classifiers made the system user-friendly as well as more robust and adaptive against variation in illumination and camera optics. In order to train various machine learning classifiers, the images of the µPADs were taken at 30 s and 10 min by four smartphone brands under seven different illuminations. According to the results, linear discriminant analysis exhibited the highest classification accuracy (98.7%) with phone-independent repeatability at t = 30 s and the accuracy was preserved for 10 min. The proposed system also showed excellent selectivity in the presence of various interfering molecules and good detection performance in tap water. [ABSTRACT FROM AUTHOR]

Learning Classifier Systems (LCSs) are a group of rule-based evolutionary computation techniques, which have been frequently applied to data mining tasks. The LCSs' rules are designed to be human-readable to enable the underlying knowledge to be investigated. However, the models for the majority of domains with high feature interaction contain a large number of rules that cooperatively represent the knowledge. However, the interaction between many rules is too complex to be comprehended by humans. Thus, it is hypothesized that translating the models' underlying patterns into human-discernable visualizations will advance the understanding of the learned patterns and LCSs themselves. Interrogatable artificial Boolean domains with varying numbers of attributes are considered as benchmarks. Three new visualization techniques, termed as Feature Importance Map, Action-based Feature Importance Map, and Action-based Feature's Average value Map, successfully produce interpretable results for all the complex domains tested. This includes both tracing the training progress and analyzing the trained models from LCSs. The visualization techniques' ability to handle complex optimal solutions is observed for the 14-bits Majority-On problem, where the patterns from 6435 different cooperating rules were translated into human-discernable graphs. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL) has become a highly successful framework for learning in Markov decision processes (MDP). Due to the adoption of RL in realistic and complex environments, solution robustness becomes an increasingly important aspect of RL deployment. Nevertheless, current RL algorithms struggle with robustness to uncertainty, disturbances, or structural changes in the environment. We survey the literature on robust approaches to reinforcement learning and categorize these methods in four different ways: (i) Transition robust designs account for uncertainties in the system dynamics by manipulating the transition probabilities between states; (ii) Disturbance robust designs leverage external forces to model uncertainty in the system behavior; (iii) Action robust designs redirect transitions of the system by corrupting an agent's output; (iv) Observation robust designs exploit or distort the perceived system state of the policy. Each of these robust designs alters a different aspect of the MDP. Additionally, we address the connection of robustness to the risk-based and entropy-regularized RL formulations. The resulting survey covers all fundamental concepts underlying the approaches to robust reinforcement learning and their recent advances. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL) allows an agent to solve sequential decision-making problems by interacting with an environment in a trial-and-error fashion. When these environments are very complex, pure random exploration of possible solutions often fails, or is very sample inefficient, requiring an unreasonable amount of interaction with the environment. Hierarchical reinforcement learning (HRL) utilizes forms of temporal- and state-abstractions in order to tackle these challenges, while simultaneously paving the road for behavior reuse and increased interpretability of RL systems. In this survey paper we first introduce a selection of problem-specific approaches, which provided insight in how to utilize often handcrafted abstractions in specific task settings. We then introduce the Options framework, which provides a more generic approach, allowing abstractions to be discovered and learned semi-automatically. Afterwards we introduce the goal-conditional approach, which allows sub-behaviors to be embedded in a continuous space. In order to further advance the development of HRL agents, capable of simultaneously learning abstractions and how to use them, solely from interaction with complex high dimensional environments, we also identify a set of promising research directions. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is one of the most exciting fields of computing today. In the experimental evaluation, they showed that the new grammatical inference algorithm gives the best results in comparison to other automata or grammar learning methods as well as ML approach combining an unsupervised data-driven distributed representation and SVM. In [[11]], this problem is solved using several ML algorithms based on different student data including individual course grades and grade point averages. ML covers a wide range of learning algorithms, including classical ones such as linear regression, k-nearest neighbors, decision trees, support vector machines and neural networks, and newly developed algorithms such as deep learning and boosted tree models. [Extracted from the article]

Autoencoders are data-specific compression algorithms learned automatically from examples. The predominant approach has been to construct single large global models that cover the domain. However, training and evaluating models of increasing size comes at the price of additional time and computational cost. Conditional computation, sparsity, and model pruning techniques can reduce these costs while maintaining performance. Learning classifier systems (LCSs) are a framework for adaptively subdividing input spaces into an ensemble of simpler local approximations that together cover the domain. LCS perform conditional computation through the use of a population of individual gating/guarding components, each associated with a local approximation. This article explores the use of an LCS to adaptively decompose the input domain into a collection of small autoencoders, where local solutions of different complexity may emerge. In addition to the benefits in convergence time and computational cost, it is shown possible to reduce the code size as well as the resulting decoder computational cost when compared with the global model equivalent. [ABSTRACT FROM AUTHOR]

This paper investigates the joint relay and channel selection problem using a deep reinforcement learning (DRL) algorithm for cooperative communications in a dynamic jamming environment. The latest types of jammers include the mobile and smart jammer that contains multiple jamming patterns. This new type of jammer poses serious challenges to reliable communications such as huge environment states, tightly coupled joint action selections and real‐time decision requirements. To cope with these challenges, a DRL‐based relay‐assisted cooperative communication scheme is proposed. In this scheme, the joint selection problem is constructed as a Markov decision process (MDP) and a double deep Q network (DDQN) based anti‐jamming scheme is proposed to address the unknown and dynamic jamming behaviors. Concretely, a joint decision‐making network composed of three sub‐networks is designed and the independent learning method of each sub‐network is proposed. The simulation results show that the user agent is able to anticipate the jammer behaviors and elude the jamming in advance. Furthermore, compared with the sensing‐based algorithm, the Q learning‐based algorithm and the existing DRL‐based anti‐jamming approaches, the proposed algorithm maintains a higher average normalized throughput. [ABSTRACT FROM AUTHOR]

Development without any defect is unsubstantial. Timely detection of software defects favors the proper resource utilization saving time, effort and money. With the increasing size and complexity of software, demand for accurate and efficient prediction models is increasing. Recently, search-based techniques (SBTs) have fascinated many researchers for Software Defect Prediction (SDP). The goal of this study is to conduct an empirical evaluation to assess the applicability of SBTs for predicting software defects in object-oriented (OO) softwares. In this study, 16 SBTs are exploited to build defect prediction models for 13 OO software projects. Stable performance measures — GMean, Balance and Receiver Operating Characteristic-Area Under Curve (ROC-AUC) are employed to probe into the predictive capability of developed models, taking into consideration the imbalanced nature of software datasets. Proper measures are taken to handle the stochastic behavior of SBTs. The significance of results is statistically validated using the Friedman test complied with Wilcoxon post hoc analysis. The results confirm that software defects can be detected in the early phases of software development with help of SBTs. This paper identifies the effective subset of SBTs that will aid software practitioners to timely detect the probable software defects, therefore, saving resources and bringing up good quality softwares. Eight SBTs — sUpervised Classification System (UCS), Bioinformatics-oriented hierarchical evolutionary learning (BIOHEL), CHC, Genetic Algorithm-based Classifier System with Adaptive Discretization Intervals (GA_ADI), Genetic Algorithm-based Classifier System with Intervalar Rule (GA_INT), Memetic Pittsburgh Learning Classifier System (MPLCS), Population-Based Incremental Learning (PBIL) and Steady-State Genetic Algorithm for Instance Selection (SGA) are found to be statistically good defect predictors. [ABSTRACT FROM AUTHOR]

The use of advanced machine learning (ML) models for manufacturing could potentially reduce the pre-production testing and validation time for new processes. Once we decide that ML is indeed a suitable tool to apply in smart manufacturing processes, the challenge lies in training, validating, and testing an ML model in a pre-production environment so that engineers can be confident that the model building effort can be successfully transitioned to actual production. This paper aims at explaining the in-works of a given in-situ classifier for predicting the quality welds in ultrasonic welded battery tabs. Predicting the quality of new samples cannot attain full certainty due to characteristics of the data the model was trained on (e.g., noisy or wrongly labeled). By developing explainable methods to such connectionist learning models (also known as black boxes), we show why the classifier outputs were predicted, making these predictions better understood and trustworthy. [ABSTRACT FROM AUTHOR]

Evolutionary machine learning research aims to develop classifier systems that can solve complex and hard tasks. This paper addresses the problem of inferring a regular expression from a given set of strings for automating the task of information extraction. To the best of our knowledge, this paper is the first to propose the extension of accuracy-based classifier system XCS to learn the regular expressions for text extraction. This new system named as XCSREA includes tree-like code fragments to learn regular expressions. The genetic algorithm in action sets uses two-point crossover with uniform mutation and Roulette wheel parent selection method. Seven different datasets, each with three different lengths, are used to compare the performance of the proposed model with standard genetic programming (GP) approach. The experimental results demonstrate that XCSREA outperforms standard GP approach when sufficiently large numbers of classifiers are used. [ABSTRACT FROM AUTHOR]

Human beings follow a continuous learning paradigm, i.e., they learn to solve smaller and relatively easy problems, retain the learnt knowledge and apply that knowledge to learn and solve more complex and large-scale problems of the domain. Currently, most machine learning and evolutionary computing systems lack this ability to reuse the previous learnt knowledge. This paper presents a lifelong machine learning model for text classification that extracts the useful knowledge from simple problems of a domain and reuses the learnt knowledge to learn complex problems of the domain. The proposed approach adopts a rule-based learning classifier system, and a rich encoding scheme is used to extract and reuse building units of knowledge. The experimental results show that the continuous learning approach outperformed the baseline classifier system. [ABSTRACT FROM AUTHOR]

A learning classifier system is an adaptive system that obtains a set of appropriate action rules that adapts to multistep problems by training action rules defined in if‐then form by trial and error process, in a similar framework as reinforcement learning. Because of that the input signals of the classifier system are encoded into binary values, bit strings are often lengthened when dealing with such a problem that the state of the environment continuously changes. A neural network can treat with real values as input signal; however, it cannot be applied to multistep problems. This paper proposes a system that responds to problems such that the state of the environment continuously changes by combining a neural network and a classifier system, and actions are selected from multiple options, so that output can be defined as discrete values. In order to verify the effectiveness of the proposed system, this paper conducts several numerical experiments using benchmarks corresponding to multistep problems defined by continuous values. [ABSTRACT FROM AUTHOR]

Comprehending why an artificial intelligence–based agent makes certain predictions is often discussed as one of the central issues that needs to be solved in the near future to make such agents ubiquitously deployable. However, many common black-box machine learning models are hard to analyse. Recently, we proposed a new machine learning algorithm to construct concise sets of rules. This algorithm, SupRB, creates compact, interpretable and transparent models. One key feature that differentiates it from many similar previously proposed systems is that it optimizes the model selection tasks concerning rule discovery and rule set composition separately. Therefore, users can tailor a model's structure more easily towards fulfilling use-case specific explainability requirements. Keeping the model selection tasks separate during optimization allows for more clearly defined goals when compared to state-of-the-art systems and – importantly – facilitates independent rule fitnesses, making the training process easier to explain and the models easier to analyse post-hoc. In this paper, we extend previous benchmarking of SupRB, where it was compared with XCSF, a prominent rule set learning system, to also include Decision Trees (DTs) and Random Forests (RFs) in our analysis. We find that – as expected – SupRB performs comparable to XCSF, while allowing easier control of model structure and showing a substantially smaller sensitivity to random seeds and data splits, and it performs slightly worse than RFs, and better than DTs. Importantly, the overall model size was substantially smaller with SupRB, making it the easiest to interpret system while not falling significantly short in terms of prediction errors. • Detailed the presentation of SupRB, an interpretable machine learning algorithm. • SupRB performs competitively on errors with generally smaller models. • Analysing a rule is easy to do even for non-experts. [ABSTRACT FROM AUTHOR]

Highlights • An integrated scheme is proposed for prognostics of Lithium-ion batteries. • The prognostic scheme contains pre-processing and prediction modules. • Novel imputation techniques handle battery data with missing observations. • The proposed multiple imputation technique reveals the uncertainty of estimations. • Extreme learning machines predict the remaining useful life over a long horizon. Abstract This paper focuses on the development of a prognostic scheme for estimating the remaining useful life (RUL) of Lithium-ion batteries with missing observations. The scheme has two main modules based on extreme learning machines: pre-processing and prediction. The pre-processing module uses novel single and multiple imputation techniques to estimate the missing observations. The prediction module aims to obtain precise predictions even in the presence of missing observations and with the related uncertainty. The pre-processing module sends imputed subsets of samples to the prediction module, which makes use of extreme learning machines for one-step and multi-steps predictions. The prediction module contains various multi-steps prediction strategies including iterative, direct and DirRec, which use the constant-current experimental capacity data for the long-term prediction of the remaining useful life. Accurate prediction of RUL requires continuity in the time-series dataset. The proposed scheme is designed to build an intelligent prediction system with the ability to handle time-series data containing missing values and is robust enough to generate a complete time-series dataset and, then, make short or long term predictions. The experimental results confirm that the proposed framework can be beneficial for intelligent diagnostic and prognostic systems related to battery as well as other wide range of applications. The main focus of the paper is the development of an integrated imputation-prediction scheme and not the evaluation of individual performances of the imputation or prediction techniques. [ABSTRACT FROM AUTHOR]

Reinforcement learning has great potential in solving practical problems, but when combining it with neural networks to solve small scale discrete space problems, it may easily trap in a local minimum value. Traditional reinforcement learning utilizes continuous updating of a single agent to learn policies, which easily leads to a slow convergence speed. In order to solve the above problems, we combine asynchronous methods with existing tabular reinforcement learning algorithms, propose a parallel architecture to solve the discrete space path planning problem, and present some new variants of asynchronous reinforcement learning algorithms. We apply these algorithms on the standard reinforcement learning environment problems, and the experimental results show that these methods can solve discrete space path planning problems efficiently. One of these algorithms, Asynchronous Phased Dyna-Q, which surpasses existing asynchronous reinforcement learning algorithms, can well balance exploration and exploitation. [ABSTRACT FROM AUTHOR]

Sentiment analysis of public views and spam detection from social media text messages are two challenging data analysis tasks due to short informal text. This paper investigates the performance of learning classifier systems (LCS), which are rule-based machine learning techniques, in sentiment analysis of twitter messages and movie reviews, and spam detection from SMS and email data sets. In this study, an existing LCS technique is extended by introducing a novel encoding scheme to represent classifier rules in order to handle the sparseness in feature vectors, which are generated using the term frequency inverse document frequency of word n-grams and sentiment lexicons. The obtained results show that the proposed encoding scheme smoothed the learning process and generated consistently good results in all experiments conducted in this study. [ABSTRACT FROM AUTHOR]

In this study we apply two methods for data collection that are relatively new in the field of atmospheric science. The two developed methods are designed to collect essential geo-localized information to be used as input data for a high resolution emission inventory for residential wood combustion (RWC). The first method is a webcrawler that extracts openly online available real estate data in a systematic way, and thereafter structures them for analysis. The webcrawler reads online Norwegian real estate advertisements and it collects the geo-position of the dwellings. Dwellings are classified according to the type (e.g., apartment, detached house) they belong to and the heating systems they are equipped with. The second method is a model trained for image recognition and classification based on machine learning techniques. The images from the real estate advertisements are collected and processed to identify wood burning installations, which are automatically classified according to the three classes used in official statistics, i.e., open fireplaces, stoves produced before 1998 and stoves produced after 1998. The model recognizes and classifies the wood appliances with a precision of 81%, 85% and 91% for open fireplaces, old stoves and new stoves, respectively. Emission factors are heavily dependent on technology and this information is therefore essential for determining accurate emissions. The collected data are compared with existing information from the statistical register at county and national level in Norway. The comparison shows good agreement for the proportion of residential heating systems between the webcrawled data and the official statistics. The high resolution and level of detail of the extracted data show the value of open data to improve emission inventories. With the increased amount and availability of data, the techniques presented here add significant value to emission accuracy and potential applications should also be considered across all emission sectors. [ABSTRACT FROM AUTHOR]

In this paper, a rank-based nonparametric statistical test for measuring the effect of cooperation between optimization agents solving the multi-mode resource-constrained project scheduling problem is presented. To solve this NP-hard optimization problem, different methods are applied including population- and agent-based approaches. One of them is a team of asynchronous agents composed of multiple optimization agents, management agents, and common memories, which through interactions produce solutions of hard optimization problems. Optimization agents represent different methods including local search, path relinking, or tabu search. Interactions are managed through various cooperation strategies based on applying heuristics, reinforcement learning, or population learning. [ABSTRACT FROM AUTHOR]

The bark is a very distinctive vocalization of the dog. It is very common and a mean for interaction with humans. However, the scope of their automated analysis by computational techniques, as well as the possible applications to which they can give rise have been little explored. In this study, we describe the process to develop an automatic classifier that can identify individual dogs based on their barks. We created a database with more than 6,000 barks applying positive and negative stimuli to dogs. We acoustically characterized the barking samples using a signal processing tool that extracts large sets of features. Based on these sets, we generated optimal subsets of features to feed machine learning algorithms which trained classification models. We evaluated such models and compared the classification performance of different algorithms. We analyzed the pertinence of training specific models per each breed. The classification obtained outperform the results previously reported in similar works. Our findings suggest that practical applications could be constructed on this kind of technology. [ABSTRACT FROM AUTHOR]

Extreme Learning Machines (ELM) have been successfully applied to variety of classification problems by utilizing a single descriptor type. However, a single descriptor may be insufficient for the visual classification task, due to the high level of intra-class variability coupled with low inter-class distance. Although several studies have investigated methods for combining multiple descriptors by ELM, they predominantly apply a simple concatenation of descriptors before classifying them. This type of descriptor fusion may impose problems of descriptor compatibility, high dimensionality and restricted accuracy. In this paper, we propose a hierarchical descriptors fusion strategy at the decision level (“late-fusion”), which relies on ELM ensembles (ELM-E). The proposed method, denoted as H-ELM-E, effectively combines multiple complementary descriptors by a two-level ELM-E based architecture, which ensures that a more informative descriptors will gain more impact on the final decision. In the first level, a separate ELM-E classifier is trained for every image descriptor. In the second level, the output scores from the previous level are aggregated into the mid-level representation which is conducted to an additional ELM-E classifier. The exhaustive experimental evaluation confirmed that the proposed hierarchical ELM-E based strategy is superior to the single-descriptor methods as well as “early fusion” of multiple descriptors, for the visual classification task. Additionally, it was shown that significant accuracy improvement is achieved by integrating ensembles of ELM as a basic classifier, instead of using a single ELM. [ABSTRACT FROM AUTHOR]

Multiple Classifier Systems (MCS) have been widely studied as an alternative for increasing accuracy in pattern recognition. One of the most promising MCS approaches is Dynamic Selection (DS), in which the base classifiers are selected on the fly, according to each new sample to be classified. This paper provides a review of the DS techniques proposed in the literature from a theoretical and empirical point of view. We propose an updated taxonomy based on the main characteristics found in a dynamic selection system: (1) The methodology used to define a local region for the estimation of the local competence of the base classifiers; (2) The source of information used to estimate the level of competence of the base classifiers, such as local accuracy, oracle, ranking and probabilistic models, and (3) The selection approach, which determines whether a single or an ensemble of classifiers is selected. We categorize the main dynamic selection techniques in the DS literature based on the proposed taxonomy. We also conduct an extensive experimental analysis, considering a total of 18 state-of-the-art dynamic selection techniques, as well as static ensemble combination and single classification models. To date, this is the first analysis comparing all the key DS techniques under the same experimental protocol. Furthermore, we also present several perspectives and open research questions that can be used as a guide for future works in this domain. [ABSTRACT FROM AUTHOR]

This paper describes a framework for Dynamic Classifier Selection (DCS) whose novelty resides in its use of features that address the difficulty posed by the classification problem in terms of orienting both pool generation and classifier selection. The classification difficulty is described by meta-features estimated from problem data using complexity measures. Firstly, these features are used to drive the classifier pool generation expecting a better coverage of the problem space, and then, a dynamic classifier selection based on similar features estimates the ability of the classifiers to deal with the test instance. The rationale here is to dynamically select a classifier trained on a subproblem (training subset) having a similar level of difficulty as that observed in the neighborhood of the test instance defined in a validation set. A robust experimental protocol based on 30 datasets, and considering 20 replications, has confirmed that a better understanding of the classification problem difficulty may positively impact the performance of a DCS. For the pool generation method, it was observed that in 126 of 180 experiments (70.0%) adopting the proposed pool generator allowed an improvement of the accuracy of the evaluated DCS methods. In addition, the main results from the proposed framework, in which pool generation and classifier selection are both based on problem difficulty features, are very promising. In 165 of 180 experiments (91.6%), it was also observed that the proposed DCS framework based on the problem difficulty achieved a better classification accuracy when compared to 6 well known DCS methods in the literature. [ABSTRACT FROM AUTHOR]

Multilabel classification (MLC) learning, which is widely applied in real-world applications, is a very important problem in machine learning. Some studies show that a clustering-based MLC framework performs effectively compared to a nonclustering framework. In this paper, we explore the clustering-based MLC problem. Multilabel feature selection also plays an important role in classification learning because many redundant and irrelevant features can degrade performance and a good feature selection algorithm can reduce computational complexity and improve classification accuracy. In this study, we consider feature dependence and feature interaction simultaneously, and we propose a multilabel feature selection algorithm as a preprocessing stage before MLC. Typically, existing cluster-based MLC frameworks employ a hard cluster method. In practice, the instances of multilabel datasets are distinguished in a single cluster by such frameworks; however, the overlapping nature of multilabel instances is such that, in real-life applications, instances may not belong to only a single class. Therefore, we propose a MLC model that combines feature selection with an overlapping clustering algorithm. Experimental results demonstrate that various clustering algorithms show different performance for MLC, and the proposed overlapping clustering-based MLC model may be more suitable. [ABSTRACT FROM AUTHOR]