The integration of artificial intelligence (AI) in healthcare decision-making has revolutionised the diagnosis and treatment of many diseases. However, challenges such as model interpretability, data quality, algorithmic bias, and ethical considerations remain a barrier. This paper presents a multi-algorithm approach for heart disease diagnosis that prioritises accuracy, explainability, and ethical AI principles. It also aligns with Explainable Artificial Intelligence (XAI) principles by highlighting ante-hoc transparency through careful feature selection and a tailored CNN model design for heart disease diagnosis. By leveraging interpretable AI techniques and addressing key challenges, this paper demonstrates how trustworthy and reliable AI systems can transform healthcare. Additionally, it explores the potential of post-hoc explainability techniques, such as SHAP and LIME, to clarify the model decisions and build trust among the healthcare professionals. This work bridges the gap between AI and the clinical practice.

Monitoring and forecasting the progress of information technology projects stands as a significant challenge in project management. Over the past two decades, agile project management has become a crucial factor influencing project success. Despite this, existing research has not presented a comprehensive model capable of predicting project outcomes based on agile features. In light of this, this study aims to develop a predictive model for information technology project outcomes using agility metrics. The results indicate that metrics related to teamwork and the team's capabilities, along with their collective experience, have the most significant impact on project success. The study employs the Decision Tree as an interpretable model to establish rules and predict project success. The accuracy of the model designed in this study is an impressive 97%, surpassing the accuracy of SVM at 71 % and KNN at 82%.

A new portable solution is proposed based on Convolutional Neural Networks (CNN) to increase the speed and accuracy of detecting violence behaviour on edge devices. This solution has numerous applications in public safety. A combination of surveillance using CCTV cameras and Unmanned Aerial Vehicles (UAVs) is used to demonstrate the real-world surveillance use cases to monitor abnormal behaviors in public. The proposed solution delivers 95.01% accuracy while taking 13.2ms for inference on GeForce GTX 1660 Ti GPU and reaching 38 frames per second throughput on Jetson AGX Orin measured on a combination of Drone-action and chu-surveillance-violence-detection datasets. The results show the strong practical application potential of the proposed solution in terms of real-time performance, visual quality, and high accuracy.

The rapid expansion of the AI market has outpaced the development of adequate regulations and guidelines, resulting in deficiencies in tools for practitioners and stakeholders in AI systems (AIS). These guidelines are re-evaluated and re-scoped to emphasise positive ethical design and supporting operationalisation, resulting in an actionable, concrete, and accessible toolkit. The work aims to contribute to the optimisation of SWI-Prolog fuzzy systems extended over a new problem domain. The work further introduces the integration of Explainable AI (XAI) principles within the proposed ethical framework as an extension of current XAI research. A novel fuzzy expert system is presented with a proposed system algorithm, employing SWI-Prolog extended through the Constraint Logic Programming library for membership function constraint propagation. A Command Line Interface (CLI) is presented for user interaction, with functionality to allow users to store and load data locally.

This research addresses the critical issue of fairness bias in machine learning predictions and explanations around the Nigerian Polytechnic admission process. We propose a novel approach that leverages Structural Causal Models (SCMs) to design an ontological framework for detecting bias and adjusting features in Local Interpretable Model-agnostic Explanations (LIME). The SCM ontology provides a principled approach to identify the features that contribute to fairness bias, and we show the presence of bias in LIME explanations for the Polytechnic admission dataset. To mitigate this bias, we propose an ablation technique for feature adjustment in a fair-LIME framework that uses SCM ontology. Experimental results on the Benpoly admission dataset show a remarkable reduction in fairness bias with high fidelity to the predictions of the black-box model. The fair-LIME explanations provided a valid and unbiased interpretation of the factors driving admission decisions. In this regard, the paper significantly extends the scope of eXplainable AI by developing a principled method for obtaining an unbiased explanability aimed at fostering the growth of trustworthy AI systems. The fair-LIME framework holds strong implications towards making the development of transparent, accountable, and ethical AI systems a reality for high-stakes decision-making processes.

This paper explores the application of machine learning (ML) and explainable AI (XAI) techniques for detecting money laundering in financial transactions. A novel approach is introduced that combines a deep neural network (DNN) with SHapley Additive exPlanations (SHAP) to enhance the transparency and effectiveness of anti-money laundering (AML) systems. The proposed model demonstrates superior performance over benchmark models, achieving high precision (0.994585), recall (0.994500), F1 score (0.994551), and ROC AUC (0.994525) in identifying fraudulent transactions using a synthetic dataset derived from real financial logs. Through a global explainability analysis, key indicators of fraudulent activities, such as high transaction amounts and prolonged transaction durations, are identified. This study contributes to the AML field by improving model accuracy and providing insights into the decision-making processes of complex ML models. Future research will focus on applying local explanations and utilizing larger real world datasets to further enhance model performance and interpretability.

The study proposes using five benchmark machine-learning models alongside XGBoost, applied for the first time to an existing case study to predict the success of suicide of terrorist attacks. Utilizing data from the Global Terrorism Database (GTD), the study evaluates model effectiveness to aid decision-making for emergency responders and policymakers. Employing explainable Artificial Intelligence (XAI) models like SHAP ensures transparent decision-making processes. XGBoost performed best for accuracy and performance, while LightGBM excelled in explainability, with SHAP providing global and local insights into their decision-making. The primary goal is to enhance user comprehension and facilitate informed decision-making in critical scenarios, prioritizing transparency, and trustworthiness.

Phishing threats have remained a long-standing information security issue for many years, causing billions of pounds in losses both in the United Kingdom and worldwide [1]. The aim of the study was to develop and evaluate the performance of the Machine Learning models that would detect and monitor phishing attacks more accurately. A single dataset with 42 features, a total of 247950 phishing and non-phishing emails was used to develop eight supervised machine learning models. The metrics used in evaluating the models show that the enhanced hybrid algorithm developed from combining two models (decision trees and the random forests) from the trained models generated the best classifier with an accuracy of 96%, precision 98%, f-measure 96%, sensitivity 94%, MCC 92% and ROC 96%. The enhanced hybrid voting model developed was integrated with a Django web application using 13 important features to build an accurate phishing detection and monitoring application. The model is proposed as a novel hybrid model because it demonstrated higher classification capabilities due to its inherent design to deal with complex patterns, overfitting issue and the presence of many features when compared to the other single analysis models in the experiments.

This paper presents the first published application of multiple existing machine learning methods to a subset of features taken from the Profiles of Individual Radicalization in the United States (PIRUS) database to predict the feature ‘violent’. The best-performing model in terms of accuracy is the Hist Gradient Boosting model, with an accuracy of 89.06%, which is an improvement of more than 2.5% compared to the benchmark application. Permutation Feature Importance (PFI) and the explanation framework SHAP were then applied to explain the model predictions. Using both of these techniques together allows for a holistic view of both the model's inner workings and the impact of the features on the results.

This paper presents an application of a knowledge extraction method developed by the authors to animal farm management. The main purpose of the paper is to process information on the relationships between profits and costs in livestock farming. The paper uses expert knowledge as a benchmark approach for knowledge extraction and a fuzzy method for comparative validation. The academic novelty in the paper is the first-time application of the relation based knowledge extraction method to livestock farming.

We present fuzzy boundaries, a method of segmenting text whereby we consider a population of boundaries to be in a fuzzy-state. Furthermore, we aim to use this method to present a multifaceted segmentation approach that is applicable across various domains that require the use of text segmentation: Text summerisation, Rhetorical Stricture Theory, sentence-based segmentation, paragraph-based segmentation and search-algorithms that require topic-based querying. The work first outlines the subject domain together with previous work surrounding the topic of text segmentation. We move to discussing the rationale, exploring our motivations for such a method. The model and its composition is then described next along side the direction taken in the implementation of the model. We move to the results and performance of our method concluding finally with a discussion on the benefits and justifications for our propositions.

Sentiment Analysis is a strand of Natural Language Processing that deals with the emotional polarity a given piece of text has. To gain this understanding from just a string of words, we must first consider a suitable way to break down the text to further classify what each part means. This can be done in a plethora of ways, which mostly stem from the understanding of a classifier. We believe that there is a large amount of information stored in structure-based features within the text, for instance, where the writer may place negation-terms not, neither and how this affects the overall polarity. Similarly, how words in sentences or sections, remote to the current-analysed section, may affect the polarity of said section. A combination of features from both a conventional and a structure-based understanding may also provide us with a larger accuracy in polarity. Therefore, this paper aims to explain both conventional sentiment analysis methods with structure-based methods as well as their practices, advantages and disadvantages concluding with how sentiment analysis can move forward with the appropriation of hybrid methods (methods involving motifs, practices and understandings) from conventional and structure-based methods, for classification.

The concept of q-rung orthopair fuzzy set (q-ROFS) is the extension of intuitionistic fuzzy set (IFS) in which the sum of the qth power of the support for and the qth power of the support against is bounded by one. Therefore, the q-ROFSs are an important way to express uncertain information in broader space, and they are superior to the IFSs and the Pythagorean fuzzy sets. In this paper, the familiarity degree of the experts with the evaluated objects is incorporated to the initial assessments under q-rung orthopair fuzzy environment. For this, some aggregation operators are proposed to combine these two types of information. Their some important properties are also well proved. Furthermore, these developed operators are utilized in a multicriteria decision-making approach and demonstrated with a real life problem of customers' choice. Then, the experimental results are compared with other existing methods to show its superiority over recent research works.

The papers in this special section focus on recent developments and emerging topics in the area of deep fuzzy models that address some of the problems and limitations above. These models have been known under different names, such as hierarchical fuzzy systems and fuzzy networks. They are usually well suited for performing multiple functional compositions at either crisp or linguistic level. Deep learning has gained significant attention within the computational intelligence community in recent years. Its success has been mainly due to the increased power of modern computational platforms in terms of their ability to collect, store, and process large volumes of data. This has led to a substantial increase in the effectiveness and efficiency of data management. As a result, it has become possible to achieve high accuracy for some benchmark learning tasks, such as object classification and image recognition within a short time frame. The most common implementation of deep learning has been through neural networks due to the ability of their layers of neurons to perform multiple functional compositions as part of a multistage learning process.

In this article, the fuzzy property is described by means of the Z-number as the coefficients and variables of the fuzzy equations. This alteration for the fuzzy equation is appropriate for system modeling with Z-number parameters. In this article, the fuzzy equation with Z-number coefficients and variables is tended to be used as the models for the uncertain systems. The modeling issue related to the uncertain system is to obtain the Z-number coefficients and variables of the fuzzy equation. Nevertheless, it is extremely hard to get the Z-number coefficients of the fuzzy equations. In this article, in order to model the uncertain nonlinear systems, a novel structure of the multilayer neural network is utilized in such a manner that it is able to obtain the Z-number coefficients of the fuzzy equation. The suggested technique is validated by some examples with applications.

For a long time technicians and engineers have used geometric changes of objects for the purpose of enhancement of heat transfer. The discovery and use of nanofluids and their unique properties lead to a new revolution on the heat transfer. This paper presents the simulation of Ansis software applied to the flow tube with a constant flux, also studies the effect of baffles and the use of nano particles on heat transfer.

Existing Side-channel attack techniques, such as meltdown attacks, show that attackers can exploit the microarchitecture and OS vulnerabilities to achieve their goals. In this paper, we present the development of our real-time system for detecting side-channel attacks. Unlike previous works, our proposed detection system does not rely on synchronisation between the attackers and victims. Instead, it uses processors' performance indicators to capture malicious Flush+ Reload activities with an accuracy of up to 99%. Moreover, the detection activities can be achieved with minimum time delay in both native and cloud systems with a low overhead performance of approximately less than 1% in the host system.

Software metrics based maintainability prediction is leading to development of new sophisticated techniques to construct prediction models. This paper proposes a new software maintainability prediction framework, which bases on Fuzzy Network, a novel exploratory modeling technique. The proposed framework utilizes both the metric data collected from software system and the subjective appraisals from experts. An application example of the framework is shown. In comparison to the Standard Fuzzy System based models, Fuzzy Network based models improves the transparency more than 71.3% and the accuracy more than 11.0%. It is confirmed that Fuzzy Network based framework is more appropriate for constructing SMP model.

This paper presents fuzzy similarity based Fuzzy Technique for Order Performance by Similarity to Ideal Solution (TOPSIS) for z-numbers. The classical fuzzy TOPSIS techniques use closeness coefficient to determine the rank order by calculating Fuzzy Positive Ideal Solution (FPIS) and Fuzzy Negative Ideal Solution (FNIS) simultaneously. The authors propose fuzzy similarity to replace closeness coefficient by doing ranking evaluation. Fuzzy similarity is used to calculate the similarity between two fuzzy ratings (FPIS and FNIS). Fuzziness is not sufficient enough when dealing with real information and a degree of reliability of the information is very critical. Hence, the implementation of z-numbers is taken into consideration as they can capture better the knowledge of human being and are extensively used in uncertain information development to deal with linguistic decision making problems. A numerical example is given to illustrate the application of the proposed technique in ranking company performance assessment. The results show that it is highly feasible to use the proposed technique in performance assessment.

This paper presents a novel extension of the Technique for Ordering of Preference by Similarity to Ideal Solution (TOPSIS) method. The method is based on aggregation of rules with different linguistic values of the output of fuzzy networks to solve multi criteria decision-making problems whereby both benefit and cost criteria are presented as subsystems. Thus the decision maker evaluates the performance of each alternative for decision process and further observes the performance for both benefit and cost criteria. The aggregation of rule bases in a fuzzy system maps the fuzzy membership functions for all rules to an aggregated fuzzy membership function representing the overall output for the rules. This approach improves significantly the transparency of the TOPSIS methods, while ensuring high effectiveness in comparison to established approaches. To ensure practicality and effectiveness, the proposed method is further tested on equity selection problems. The ranking produced by the method is comparatively validated using Spearman rho rank correlation. The results show that the proposed method outperforms the existing TOPSIS approaches in terms of ranking.

Fuzzy systems consisting of networked rule bases, called fuzzy networks, capture various types of imprecision inherent in financial data and in the decision-making processes on them. This paper introduces a novel extension of the technique for ordering of preference by similarity to ideal solution (TOPSIS) method and uses fuzzy networks to solve multicriteria decision-making problems where both benefit and cost criteria are presented as subsystems. Thus, the decision maker evaluates the performance of each alternative for portfolio optimization and further observes the performance for both benefit and cost criteria. This approach improves significantly the transparency of the TOPSIS methods, while ensuring high effectiveness in comparison with established approaches. The proposed method is further tested to solve the problem of selection/ranking of traded equity covering developed and emergent financial markets. The ranking produced by the method is validated using Spearman rho rank correlation. Based on the case study, the proposed method outperforms the existing TOPSIS approaches in terms of ranking performance.

This paper introduces a novel modification of the technique for ordering of preference by similarity to ideal solution (TOPSIS) method and uses a fuzzy system with multiple rule bases to solve multi-criteria decision making problems where both benefit and cost criteria are presented as subsystems. Thus, the decision maker evaluates the performance of each alternative for optimization and further observes the performance for both benefit and cost criteria. This approach improves significantly the transparency of the TOPSIS method while ensuring high effectiveness in comparison to established methods. To ensure practicality and effectiveness of the proposed method, a traded equity case study is considered. Furthermore, the ranking based on the proposed method is validated comparatively using spearman rho correlation. The proposed method outperforms the existing TOPSIS methods in terms of ranking for the case study under consideration.

Methods are described to create more accurate sub sets of user data by introducing dead bands into data clusters. User data is collected and then mined. That produces clusters of data. Dead bands are then generated to delineate and describe the data in the clusters more accurately. This is accomplished by classifying data inside the newly created dead bands as NOT being in either of two or more clusters. For example, three clusters are generated from two. If the two were YES and NO then another set of DON'T KNOW is introduced. The new set improves the precision of choices made using data in the YES and the NO clusters. Dead bands are introduced by establishing a radius from the corners of 2-D shapes containing the clusters or by establishing a horizontal or vertical line in parallel with the edges. Each radius or edge encompasses 80% of user data nearest to the corner or edge of the data set. 20% are outside and excluded from their original set. If lines do not overlap, then a dead-band is created to contain user data that is not as confident. That increases the likelihood of accurate decisions being made about the new sets of user data. Case studies are described to demonstrate that.

Effects of motion lag on the capability of a tele-operated mobile-robot operator are investigated. Lags can occur through communication delays as tele-operated mobile robots work at a distance or because of a lack of parallel computing power as robots are enhanced with additional systems. This work concentrates on time lag in a tele-operated mobile robot system and investigates when a mobile robot operator might begin to perceive a lag in the movement of a mobile robot. A threshold of permissible lag is established for mobile robot operators that relates to the maximum time lag before an operator noticed a lag.

Due to the vast and rapid increase in data, data mining has been an increasingly important tool for the purpose of knowledge discovery to prevent the presence of rich data but poor knowledge. In this context, machine learning can be seen as a powerful approach to achieve intelligent data mining. In practice, machine learning is also an intelligent approach for predictive modelling. A special type of machine learning methods, which are known as rule based methods such as decision trees, can be used to build a rule based system as a special type of expert systems for both knowledge discovery and predictive modelling. A rule based system may be represented through different structures. The techniques for representing rules are known as rule representation, which is significant for knowledge discovery in relation to the interpretability of the model, as well as for predictive modelling with regard to efficiency in predicting unseen instances. This paper justifies the significance of rule representation. Some networked topologies for rule representation are introduced against existing techniques. The network topologies are validated using complexity analysis in order to show their advantages comparing with the existing techniques in terms of model interpretability and computational efficiency.

The lack of ability to handle vagueness in the decision making practice has been main drawback of the conventional TOPSIS. Thus, type 1, type 2 and Z fuzzy sets have been applied with conventional TOPSIS to allow experts to incorporate imperfect information in analysis. However the existing methods do not take into account the influence degree of decision makers. Hence, a novel modification of TOPSIS method to handle vagueness and imperfect information in decision making practice is presented. The concept of Z- numbers is used to present decision maker's reliability. Furthermore, a hybrid analysis of decision making process that requires the use of human sensitivity to reflect influence degree of decision maker can be often expressed by a fuzzy rule base. The ranking based on proposed method is validated comparatively using Spearmen rho correlation coefficient. The result shows proposed method outperforms the existing non rule based version of TOPSIS in terms of ranking performance.

A post processing method is described that acts on two-dimensional clusters of data produced from a data mining system. Dead bands are automatically created that further define the clusters. This was achieved by defining data within the dead bands as NOT belonging to either cluster. The three clusters produced were definitely YES, definitely NO and a new set of DON’T KNOW. The creation of the new set improved the accuracy of decisions made about the data remaining in YES and NO clusters. The introduction of the dead bands was achieved by either setting a radius during the learning process or by setting a straight line boundary. Each radius (or line) was calculated during the learning process by considering the two dimensional position of each of the users within each cluster of dimensions. A radius line (or straight line) was then introduced so that the 80% of users within a particular dimension who were nearest to the origin (or edge) were placed into a set. The other 20% were outside the radius line (or straight line) and not recorded as being part of the set. If the two lines did not overlap, then this sometimes created a dead-band that contained users with less certain results and that in turn increased the accuracy of the other sets. Two case studies are presented as examples of that improvement.

This study introduces the novel application of a fuzzy network concept to derive optimal margins for use in the treatment of cancer using external beam radiotherapy. The input data for the model is based on the effects of treatment errors, in terms of delineation, organ motion and patient set-up errors, on tumour coverage and doses to critical organs. A demonstrable improvement in the model transparency is shown by application of the fuzzy network compared to a conventional fuzzy system, whilst the model accuracy is also improved.

A novel application of a knowledge-based fuzzy logic technique was used in this study to derive margins for use in external beam radiotherapy. Radiotherapy uncertainties including set-up, delineation and organ motion-induced errors were used to calculate the changes in prostate radiobiological indices for treatment plans using stepped increments of margin sizes. The results were used to formulate the rule base and membership functions for the Sugeno type fuzzy inference system. The application of a convolution technique was found to improve the imprecision and smoothness of the original fuzzy output. Good agreement was obtained between the proposed fuzzy model margin and currently used radiotherapy margins. The new margin was applied in a prostate cancer treatment planning example and the results compared very well to current techniques.

The article considers the application of multilayer control theory in multivariable fuzzy systems. Some definitions and theorems with regard to such systems are given. Multilayer control algorithms, based on active decomposition of the original system into layers, are presented and illustrated by numerical examples. The algorithms use sets of transforming fuzzy relations, leading to a unilayer solution when added to the original fuzzy relations. The solution is dependent on a subset of state variables as the other variables are taken into account implicitly. It is shown that the number of measured variables and fuzzy relations is significantly reduced and thus real-time measurement and control implementation is faciliated. © 1997 John Wiley & Sons, Inc.