School of Computer Science and Informatics

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 20 of 3562
  • ItemEmbargo
    Securing the remote office: reducing cyber risks to remote working through regular security awareness education campaigns
    (Springer, 2024-01-29) Angafor, Giddeon Njamngang; Yevseyeva, Iryna; Maglaras, Leandros
    Cyber security threats, including risks to remote workers, are varied and diverse, with the number of scams and business email compromise breaches increasing. Firms and their staff are experiencing mass phishing attacks, several typical precursors to more sinister attacks like cyber-enabled fraud, ransomware, and denial of service (DDoS) attacks. Threat actors are leveraging new technologies such as machine learning and artificial intelligence (AI) to deliver sophisticated scam and phishing messages that are challenging for users to identify as malicious. Several businesses are increasing technical efforts in critical areas, including network hardening, robust patching, anti-malware, ransomware detection applications, and multi-factor authentication to detect, prevent, and recover from potential threats. Despite that, these measures provide only a partial solution if the users who access the systems do not have good security awareness training. In this study, we review some cyber risks related to remote working and detail how they can be remediated through regular security awareness education campaigns (SAECs). The study presents the results of a proof of concept (PoC) experiment conducted to establish the value of regular SAECs in the fight against scams and phishing attacks against remote workers. The pilot results confirm that securing the remote office requires a robust SAEC. It argues that to be successful and help staff protect business systems and data, SAECs must be regular and varied, providing opportunities for staff to understand what to look for in suspicious scams and phishing emails. Moreover, they must provide opportunities for staff to practice their knowledge and understanding through practical exercises such as spam and phishing simulation exercises, which could help users avoid falling victim to spam and phishing emails.
  • ItemMetadata only
    Sprinting into blocks: what computing, AI and gaming academics learned
    (Times Higher Education, 2024-03-13) Allman, Zoe; Coupland, Simon; Khuman, Archie; Fahy, Conor; Attwood, Luke
  • ItemOpen Access
    Evolution of Credit Scores of Enterprises in a Social Network: A Perspective Based on Opinion Dynamics
    (IEEE, 2024-02-09) Liang, Haiming; Xu, Weijun; Chiclana, Francisco; Yu, Shui; Dong, Yucheng; Herrera-Viedma, Enrique
    The use of social network to model the evolution of credit scores of networked enterprises is still a challenging task. This article develops an opinion dynamics model of the evolution of credit scores of enterprises in a social network. Firstly, based on the number of potential cooperated enterprises and the initial credit scores, the leader and follower enterprises are identified. Then, taking into consideration the cooperated benefit and discrimination cost, the cooperated utility between any two enterprises is calculated, which is used to compute the weights that one enterprise assigns to other enterprises. An opinion dynamics model on the evolution of credit scores of enterprises, inspired on the classical Friedkin–Johnsen’s social network model, is developed. Some desirable properties of the proposed opinion dynamics model are theoretically stated and proved. Finally, a numerical example is provided to illustrate the feasibility of the proposed opinion dynamics model, while a simulation analysis to investigate the joint influences of the connection probabilities and the network structure on the evolution of credit scores of enterprises is reported.
  • ItemEmbargo
    Managing heterogeneous preferences and multiple consensus behaviors with self-confidence in large-scale group decision making
    (Elsevier, 2024-02-08) Liu, Wenqi; Wu, Yuzhu; Chen, Xin; Chiclana, Francisco
    With the rapid increase of experts, groups or organizations involved in decision making, the problem of large-scale group decision making (LSGDM) has attracted increasing attention in the whole research field. Behavioral management and heterogeneous preference representation structures are two fundamental aspects of LSGDM problems. However, psychological functioning has been less considered in existing consensus models to deal with the different behavioral styles of decision-makers. Therefore, this study proposes a novel consensus reaching framework to detect and manage multiple styles of behavior in LSGDM based on heterogeneous preferences with self-confidence. Specifically, an optimization-based selection process is introduced to obtain the individual and collective preference vectors. Next, a self-confidence driven consensus approach is proposed, which includes consensus measure, clustering, detection of multiple styles of behavior, and hybrid feedback adjustment mechanism. According to the consensus level and the self-confidence level, the proposed detection of multiple styles of behavior method identifies four different behavioral subgroup types: collaborating, accommodating, competing, and avoiding. The hybrid feedback adjustment mechanism generates different feedback adjustment opinions for the four identified behavioral type subgroups. The effectiveness and characteristics of the proposed consensus approach is demonstrated with an emergency management case study and the reporting of comprehensive simulation experiments.
  • ItemOpen Access
    Learning to guide particle search for dynamic multi-objective optimization
    (IEEE, 2024-02-23) Song, Wei; Liu, Shaocong; Wang, Xinjie; Yang, Shengxiang; Jin, Yaochu
    Dynamic multiobjective optimization problems (DMOPs) are characterized by multiple objectives that change over time in varying environments. More specifically, environmental changes can be described as various dynamics. However, it is difficult for existing dynamic multiobjective algorithms (DMOAs) to handle DMOPs due to their inability to learn in different environments to guide the search. Besides, solving DMOPs is typically an online task, requiring low computational cost of a DMOA. To address the above challenges, we propose a particle search guidance network (PSGN), capable of directing individuals’ search actions, including learning target selection and acceleration coefficient control. PSGN can learn the actions that should be taken in each environment through rewarding or punishing the network by reinforcement learning. Thus, PSGN is capable of tackling DMOPs of various dynamics. Additionally, we efficiently adjust PSGN hidden nodes and update the output weights in an incremental learning way, enabling PSGN to direct particle search at a low computational cost. We compare the proposed PSGN with seven state-of-the-art algorithms, and the excellent performance of PSGN verifies that it can handle DMOPs of various dynamics in a computationally very efficient way.
  • ItemEmbargo
    Online semi-supervised active learning ensemble classification for evolving imbalanced data streams
    (Elsevier, 2024-03-04) Guo, Yinan; Pu, Jiayang; Jiao, Botao; Peng, Yanyan; Wang, Dini; Yang, Shengxiang
    Concept drift is a core challenge in classification tasks of data streams. Although many drift adaptation methods have been presented, most of them assume that labels of all data are available, which is impractical in many real-world applications. Additionally, the absence of label makes the imbalance ratio of an imbalanced data stream difficultly being obtained in time, providing the inaccurate guidance for resampling and causing poor generalization. To tackle the joint challenges, an online semi-supervised active learning method is proposed to classifier imbalanced data streams with concept drift. A newly-arrived data is first added to the sliding window, and then assigned a pseudo label in terms of its nearest cluster. Meanwhile, semi-supervised clustering algorithm offers its predicted label. Based on the above two predictive labels, cluster-based query strategy provides the criteria for the evaluation and selection of representative instances. More especially, the uncertainty and importance of instances are defined to synthetically evaluate its representativeness. After obtaining true labels of typical ones, ensemble classifier is updated by all instances in current sliding window. Experimental results on 13 synthetic and real data streams indicate that the proposed method outperforms six comparative methods on both G-mean and Recall under various labeling budgets.
  • ItemOpen Access
    BioEmoDetector: A flexible platform for detecting emotions from health narratives
    (Elsevier, 2024-02-23) Alshouha, Bashar; Serrano-Guerrero, Jesus; Chiclana, Francisco; Romero, Francisco P.; Olivas, Jose A.
    Emotion detection can play a pivotal role in healthcare. Many psychological/psychical illnesses and disorders such as depression, anxiety, or emotional crises can be explained through the study of different emotional changes suffered by a user. One of the mechanisms to detect these emotional changes is the analysis of the user’s expressions/conversations, which can be easily represented as texts. Therefore, it is necessary to count on tools able to recognize and measure the intensity of these emotions. In this sense, at the present moment biomedical and clinical pre-trained language models have been extensively used for text classification tasks; nevertheless, their applications to the field of emotion classification, specially in healthcare, remain relatively unexplored. For that reason, this paper presents BioEmoDetector, an open-source framework for emotion prediction from texts related to medical environments. This tool introduces a flexible framework leveraging multiple biomedical and clinical pre-trained language models which can work individually or together under an ensemble model. This approach provides a powerful tool for understanding the patients’ experiences through their conversations and their impact on health outcomes.
  • ItemEmbargo
    Global feedback mechanism by explicit and implicit power for group consensus in social network
    (Elsevier, 2023-12-19) Wang, Sha; Wu, Jian; Chiclana, Francisco; Ji, Feixia; Fujita, Hamido
    This paper investigates a global feedback consensus model which determines the importance of decision makers by combining explicit and implicit Power. Explicit power is obtained according to the differences in organizational hierarchy, while implicit power manifests itself by being recognized in social network interaction. Combining these two types of power, an approach is proposed to characterize the binary power of decision makers jointly to identify the key decision units. Then, a global feedback mechanism based on the binary power is investigated. Its innovation is that it searches the adjustors who are most conducive to achieve group consensus in global and implement personalized feedback services to achieve the minimum cost optimization goal. Finally, an illustrative example to verify the validity of the proposed method are reported.
  • ItemOpen Access
    Active broad learning with multi-objective evolution for data stream classification
    (Springer, 2023-08-12) Cheng, Jian; Zheng, Zhiji; Guo, Yinan; Pu, Jiayang; Yang, Shengxiang
    In a streaming environment, the characteristics and labels of instances may change over time, forming concept drifts. Previous studies on data stream learning generally assume that the true label of each instance is available or easily obtained, which is impractical in many real-world applications due to expensive time and labor costs for labeling. To address the issue, an active broad learning based on multi-objective evolutionary optimization is presented to classify non-stationary data stream. The instance newly arrived at each time step is stored to a chunk in turn. Once the chunk is full, its data distribution is compared with previous ones by fast local drift detection to seek potential concept drift. Taking diversity of instances and their relevance to new concept into account, multi-objective evolutionary algorithm is introduced to find the most valuable candidate instances. Among them, representative ones are randomly selected to query their ground-truth labels, and then update broad learning model for drift adaption. More especially, the number of representative is determined by the stability of adjacent historical chunks. Experimental results for 7 synthetic and 5 real-world datasets show that the proposed method outperforms five state-of-the-art ones on classification accuracy and labeling cost due to drift regions accurately identified and the labeling budget adaptively adjusted.
  • ItemOpen Access
    Low-carbon routing based on improved artificial bee colony algorithm for electric trackless rubber-tyred vehicles
    (Tsinghua University Press, 2023-08-02) Guo, Yinan; Huang, Yao; Ge, Shirong; Zhang, Yizhe; Jiang, Ersong; Cheng, Bin; Yang, Shengxiang
    Trackless rubber-tyerd vehicles are the core equipment for auxiliary transportation in inclined-shaft coal mines, and the rationality of their routes plays the direct impact on operation safety and energy consumption. Rich studies have been done on scheduling rubber-tyerd vehicles driven by diesel oil, however, less works are for electric trackless rubber-tyred vehicles. Furthermore, energy consumption of vehicles gives no consideration on the impact of complex roadway and traffic rules on driving, especially the limited cruising ability of electric trackless rubber-tyred vehichles (TRVs). To address this issue, an energy consumption model of an electric trackless rubber-tyred vehicle is formulated, in which the effects from total mass, speed profiles, slope of roadways, and energy management mode are all considered. Following that, a low-carbon routing model of electric trackless rubber-tyred vehicles is built to minimize the total energy consumption under the constraint of vehicle avoidance, allowable load, and endurance power. As a problem-solver, an improved artificial bee colony algorithm is put forward. More especially, an adaptive neighborhood search is designed to guide employed bees to select appropriate operator in a specific space. In order to assign onlookers to some promising food sources reasonably, their selection probability is adaptively adjusted. For a stagnant food source, a knowledge-driven initialization is developed to generate a feasible substitute. The experimental results on four real-world instances indicate that improved artificial bee colony algorithm (IABC) outperforms other comparative algorithms and the special designs in its three phases effectively avoid premature convergence and speed up convergence.
  • ItemOpen Access
    A novel fractional order variable structure multivariable grey prediction model with optimal differential background-value coefficients and its performance comparison analysis
    (Emerald, 2024-02-09) Xia, Chao; Zeng, Bo; Yang, Yingjie
    Purpose – Traditional multivariable grey prediction models define the background-value coefficients of the dependent and independent variables uniformly, ignoring the differences between their physical properties, which in turn affects the stability and reliability of the model performance. Design/methodology/approach – A novel multivariable grey prediction model is constructed with different background-value coefficients of the dependent and independent variables, and a one-to-one correspondence between the variables and the background-value coefficients to improve the smoothing effect of the background value coefficients on the sequences. Furthermore, the fractional order accumulating operator is introduced to the new model weaken the randomness of the raw sequence. The particle swarm optimization (PSO) algorithm is used to optimize the background-value coefficients and the order of the model to improve model performance. Findings – The new model structure has good variability and compatibility, which can achieve compatibility with current mainstream grey prediction models. The performance of the new model is compared and analyzed with three typical cases, and the results show that the new model outperforms the other two similar grey prediction models. Originality/value – This study has positive implications for enriching the method system of multivariable grey prediction model.
  • ItemEmbargo
    Spectrum analysis of moving average operator and construction of time-frequency hybrid sequence operator
    (Emerald, 2021-02-22) Lin, Changhai; Liu, Sifeng; Fang, Zhigeng; Yang, Yingjie
    Purpose – The purpose of this paper is to analyze the spectral characteristics of moving average operator and to propose a novel time-frequency hybrid sequence operator. Design/methodology/approach – Firstly, the complex data is converted into frequency domain data by Fourier transform. An appropriate frequency domain operator is constructed to eliminate the impact of disturbance. Then, the inverse Fourier transform transforms the frequency domain data in which the disturbance is removed, into time domain data. Finally, an appropriate moving average operator of N items is selected based on spectral characteristics to eliminate the influence of periodic factors and noise. Findings – Through the spectrum analysis of the real-time data sensed and recorded by microwave sensors, the spectral characteristics and the ranges of information, noise and shock disturbance factors in the data can be clarified. Practical implications – The real-time data analysis results for a drug component monitoring show that the hybrid sequence operator has a good effect on suppressing disturbances, periodic factors and noise implied in the data. Originality/value – Firstly, the spectral analysis of moving average operator and the novel time-frequency hybrid sequence operator were presented in this paper. For complex data, the ideal effect is difficult to achieve by applying the frequency domain operator or time domain operator alone. The more satisfactory results can be obtained by time-frequency hybrid sequence operator.
  • ItemEmbargo
    A Novel Time Series Forecasting Model for Capacity Degradation Path Prediction of Lithium-ion Battery Pack
    (Springer, 2024-01-10) Chen, Xiang; Yang, Yingjie; Sun, Jie; Deng, Yelin; Yuan, Yinnan
    Monitoring battery health is critical for electric vehicle maintenance and safety. However, existing research has limited focus on predicting capacity degradation paths for entire battery packs, representing a gap between literature and application. This paper proposes a multi-horizon time series forecasting model (MMRNet, which consists of MOSUM, flash-MUSE attention, and RNN core modules) to predict the capacity degradation paths of battery packs. First, domain knowledge (DK) extracts the features from extensive battery aging datasets. The moving sum (MOSUM) and improved flash multi-scale attention (MUSE) methods are proposed to capture capacity curve mutations and multi-scale trends. Dynamic dropout training, transposition linear architecture, residual connections, and module stacking improve model generalization and accuracy. Experiments on battery pack and cell datasets demonstrate the superior performance of MMRNet over six baseline time series models. The proposed data-driven approach effectively predicts battery degradation trajectories, with implications for condition monitoring and the safety of electric vehicles.
  • ItemOpen Access
    Machine Learning in Oil and Gas Exploration: A Review
    (IEEE, 2024-02-01) Lawal, Ahmad; Yang, Yingjie; He, Hongmei; Baisa, Nathanael L.
    A comprehensive assessment of machine learning applications is conducted to identify the developing trends for Artificial Intelligence (AI) applications in the oil and gas sector, specifically focusing on geological and geophysical exploration and reservoir characterization. Critical areas, such as seismic data processing, facies and lithofacies classification, and the prediction of essential petrophysical properties (e.g., porosity, permeability, and water saturation), are explored. Despite the vital role of these properties in resource assessment, accurate prediction remains challenging. This paper offers a detailed overview of machine learning’s involvement in seismic data processing, facies classification, and reservoir property prediction. It highlights its potential to address various oil and gas exploration challenges, including predictive modelling, classification, and clustering tasks. Furthermore, the review identifies unique barriers hindering the widespread application of machine learning in the exploration, including uncertainties in subsurface parameters, scale discrepancies, and handling temporal and spatial data complexity. It proposes potential solutions, identifies practices contributing to achieving optimal accuracy, and outlines future research directions, providing a nuanced understanding of the field’s dynamics. Adopting machine learning and robust data management methods is crucial for enhancing operational efficiency in an era marked by extensive data generation. While acknowledging the inherent limitations of these approaches, they surpass the constraints of traditional empirical and analytical methods, establishing themselves as versatile tools for addressing industrial challenges. This comprehensive review serves as an invaluable resource for researchers venturing into less-charted territories in this evolving field, offering valuable insights and guidance for future research.
  • ItemEmbargo
    Automatically constructing a health indicator for lithium-ion battery state-of-health estimation via adversarial and compound staked autoencoder
    (Elsevier, 2024-02-17) Cai, Lei; Li, Junxin; Xu, Xianfeng; Jin, Haiyan; Meng, Jinhao; Wang, Bin; Yang, Shengxiang
    Precisely assessing the state of health (SOH) has emerged as a critical approach to ensuring the safety and dependability of lithium-ion batteries. One of the primary issues faced by SOH estimate methods is their susceptibility to the influence of noise in the observed variables. Moreover, we prefer to automatically extract explicit features for data-driven methods in certain circumstances. In light of these considerations, this paper proposes an adversarial and compound stacked autoencoder for automatically constructing the SOH estimation health indicator. The compound stacked autoencoder consists of two parts. The first one is a denoising autoencoder that learns three different denoising behaviors. The second is a feature-extracting autoencoder that employs adversarial learning to improve generalization ability. The experimental results show that the proposed compound stacked autoencoder can not only get explainable explicit features but also can achieve accurate SOH estimation results compared with its rivals. In addition, the results with transfer learning demonstrate that the proposed method not only can provide high generalization ability but also be easily transferred to a new SOH estimation task.
  • ItemOpen Access
    A hybrid mode membrane computing based algorithm with applications for proton exchange membrane fuel cells
    (MDPI, 2023-07-10) Zhao, Jinhui; Zhang, Wei; Hu, Tianyu; Xu, Ouguan; Yang, Shengxiang; Zhang, Qichun
    Membrane computing is a branch of natural computing, which has been extended to solve various optimization problems. A hybrid mode membrane-computing-based algorithm (HMMCA) is proposed in this paper to solve complex unconstrained optimization problems with continuous variables. The algorithmic framework of HMMCA translates from its distributed cell-like membrane structure and communication rule. A non-deterministic evolutionary programming method and two computational rules are applied to enhance the computational performance. In a numerical simulation, 12 benchmark test functions with different variables are used to verify the algorithmic performance. The test results and comparison with three other algorithms illustrate its effectiveness and superiority. Moreover, a case study on a proton exchange membrane fuel cell (PEMFC) system parameter optimization problem is applied to validate its practicability. The results of the simulation and comparison with seven other algorithms demonstrate its practicability.
  • ItemEmbargo
    Enhancing Context Models for Point Cloud Geometry Compression with Context Feature Residuals and Multi-Loss
    (IEEE, 2024-02-13) Sun, Chang; Yuan, Hui; Li, Shuai; Lu, Xin; Hamzaoui, Raouf
    In point cloud geometry compression, context models usually use the one-hot encoding of node occupancy as the label, and the cross-entropy between the one-hot encoding and the probability distribution predicted by the context model as the loss function. However, this approach has two main weaknesses. First, the differences between contexts of different nodes are not significant, making it difficult for the context model to accurately predict the probability distribution of node occupancy. Second, as the one-hot encoding is not the actual probability distribution of node occupancy, the cross-entropy loss function is inaccurate. To address these problems, we propose a general structure that can enhance existing context models. We introduce the context feature residuals into the context model to amplify the differences between contexts. We also add a multi-layer perception branch, that uses the mean squared error between its output and node occupancy as a loss function to provide accurate gradients in backpropagation. We validate our method by showing that it can improve the performance of an octreebased model (OctAttention) and a voxel-based model (VoxelDNN) on the object point cloud datasets MPEG 8i and MVUB, as well as the LiDAR point cloud dataset SemanticKITTI.
  • ItemEmbargo
    Contextual usability of Fintech by African Caribbean Micro-Business Owners in the UK
    (Springer, 2024-05-22) Owoseni, Adebowale; Khene, Caroline; Adedigba, Adeola; Adisa, Moshood
    This research assessed in context, the usability of two widely used fintech platforms, with a focus on identifying gaps that exclude African Caribbean micro-businesses in Leicester from maximizing the affordances of the platforms. Drawing from an integrated perspective of Affordance and Activity Theories, the study involved collecting data from 12 microbusiness entrepreneurs through semi-structured usability interviews, observations, and competitive benchmarks. Data was analyzed using user journey maps, thematic analysis of observation notes, and transcribed interviews. The analysis revealed significant contextual issues that the fintech platforms have failed to address, which include community-driven needs and trust-related concerns with how data could be used for surveillance among the African-Caribbean community. Additionally, platform-specific challenges such as disparities in features across iOS and Android platforms, as well as web interfaces, posed challenges for a seamless user experience. The research also pointed out the redundancy in requesting certain information that was not relevant to all business types, highlighting the need for more tailored and context-sensitive platform designs. These findings have significant implications for inclusivity in fintech platform design as well as other digital tools that could enhance inclusivity, development, and growth of minority businesses in the UK and the Global North by extension.
  • ItemEmbargo
    Graph convolutional networks for predicting mechanical characteristics of 3D lattice structures
    (Springer, 2024-05-06) Oleka, Valentine; Zahedi, Mohsen; Taherkhani, Aboozar; Baserinia, Reza; Zahedi, Abolfazl; Yang, Shengxiang
    Recent advancements in deep learning methods encouraged researchers to apply them to process 3D objects. Initially, convolutional neural networks which have shown their ability in the processing of 2D images were used for 3D object processing. These methods need a complex process to convert 3D objects to 2D images. This conversion leads to increased computation cost and possible information loss during the transformation. This research introduces a Graph Convolutional Network approach for predicting mechanical properties of custom-designed 3D lattice structures for tissue engineering applications. Seventeen scaffold geometrics were generated for training while eight were used for testing. Unlike traditional preprocessing into images, this methodology reduces preprocessing by leveraging GCNs to directly process 3D geometrics in graph form. The experimental results show the efficiency of our proposed method in predicting 3D lattice structures.
  • ItemEmbargo
    Hand Gesture Recognition Using a Multi-modal Deep Neural Network
    (Springer, 2024-05-06) Fulsunder, Saneet; Umar, Saidu; Taherkhani, Aboozar; Liu, Chang; Yang, Shengxiang
    As devices around us get more intelligent, new ways of interacting with them are sought to improve user convenience and comfort. While gesture-controlled systems have existed for some time, they either use additional specialized imaging equipment, require unreasonable computing resources, or are simply not accurate enough to be a viable alternative. In this work, a reliable method of recognizing gestures is proposed. The built model correctly classifies hand gestures for keyboard typing based on the activity captured by an ordinary camera. Two models are initially developed for classifying video data and classifying time-series sequences of the skeleton data extracted from a video. The models use different strategies of classification and are built using lightweight architectures. The two models are the baseline models which are integrated to form a single multi-modal model with multiple inputs, i.e., video and time-series in-puts, to improve accuracy. The performances of the baseline models are then compared to the multimodal classifier. Since the multimodal classifier is based on the initial models, it naturally inherits the benefits of both baseline architectures and provides a higher testing accuracy of 100% compared to the accuracy of 85% and 75% for the baseline models respectively.