Search

This paper aims to introduce an artificial intelligence approach to enhance short-term daily streamflow forecasts through the integration of machine learning (ML) techniques with meta -heuristic (MH) algorithms. To maximize performance, a comprehensive analysis involving the autocorrelation and partial autocorrelation functions of the observed streamflow was conducted to derive ten distinct input combinations, from which the most suitable set was selected for investigation. Subsequently, six distinct ML methods, namely, decision trees, support vector regression, long short-term memory, gated recurrent unit, multi-layer perceptron, and radial basis function (as the optimal input combination), were leveraged to predict one-day-ahead streamflow. The most promising outcomes from these six explored methods were synergistically amalgamated to enhance predictive precision. This amalgamation was then optimized by utilizing parameter values hyper-tuned by four different MH algorithms. The algorithms included the shuffled frog leaping algorithm, particle swarm optimization, ant colony optimization (ACO), and gray wolf optimization (GWO). In culmination, a case study centered around the "Wick" river in the UK was conducted to rigorously evaluate the performance of the proposed method, utilizing a range of statistical evaluation indices. The results underscored the efficacy of the proposed combined method in substantially refining streamflow predictions, exhibiting an impressive level of accuracy. Additionally, the outcomes emphasized that ACO and GWO stand out as the optimal algorithms for calibrating the parameters of the amalgamated model. For instance, the R2 values attained by the combined models optimized through the ACO and GWO algorithms were 0.69422 and 0.69444, respectively. These values notably surpass those observed in the initial models, substantiating the superiority of the applied optimization approaches. [ABSTRACT FROM AUTHOR]

High-frequency financial data is more difficult to predict than low-frequency data because it possesses nonlinearity, nonstationarity, higher volatility, and long memory and is frequently accompanied by the jump phenomena. In this paper, the nonparametric regression (NR) model based on kernel function is used to fit the nonlinear relationship between the nonstationary series Y t and its lagging series to model the trend of high frequency financial time series. Furthermore, the deep learning LSTM (long short-term memory) model is applied to capture the high volatility and frequent jumps of high frequency financial data and to improve the forecasting accuracy of the residual series. The results demonstrate that the hybrid NR and LSTM model has greatly improved the forecasting accuracy in several evaluation criteria. In comparison to NR, support vector machine (SVM), LSTM, ARIMA and NR-SVM models, the mean absolute error (MAE) of NR-LSTM has reduced by 89.78%, 97.85%, 86.48%, 32.47% and 89%, respectively. In addition, we have constructed the trading strategy for the Shanghai-Shenzhen 300 index by using the NR-LSTM model. The NR-LSTM model can continue to provide good returns even during a bear market, which can serve as a guide for investors. Furthermore, the NR-LSTM model also exhibits the best forecasting effect when we model the high-frequency data of Ping An bank in China, the FTSE 100 index in the UK, and the S&P 500 index in the US. [ABSTRACT FROM AUTHOR]

Causal machine learning (ML) algorithms recover graphical structures that tell us something about cause-and-effect relationships. The causal representation provided by these algorithms enables transparency and explainability, which is necessary for decision making in critical real-world problems. Yet, causal ML has had limited impact in practice compared to associational ML. This paper investigates the challenges of causal ML with application to COVID-19 UK pandemic data. We collate data from various public sources and investigate what the various structure learning algorithms learn from these data. We explore the impact of different data formats on algorithms spanning different classes of learning, and assess the results produced by each algorithm, and groups of algorithms, in terms of graphical structure, model dimensionality, sensitivity analysis, confounding variables, predictive and interventional inference. We use these results to highlight open problems in causal structure learning and directions for future research. To facilitate future work, we make all graphs, models, data sets, and source code publicly available online. [ABSTRACT FROM AUTHOR]

Highlights • Arbitrated argumentative disputes explain outcomes of cases characterised by features and stages. • Excess features help to win the dispute and complement explanations. • High-level explanations with content-agnostic features explain passage of legislative bills. • Large-scale visualisations provide a collaborative environment for explanations. Abstract Explaining outputs determined algorithmically by machines is one of the most pressing and studied problems in Artificial Intelligence (AI) nowadays, but the equally pressing problem of using AI to explain outputs determined by humans is less studied. In this paper we advance a novel methodology integrating case-based reasoning and computational argumentation from AI to explain outcomes , determined by humans or by machines, indifferently, for cases characterised by discrete (static) features and/or (dynamic) stages. At the heart of our methodology lies the concept of arbitrated argumentative disputes between two fictitious disputants arguing, respectively, for or against a case's output in need of explanation, and where this case acts as an arbiter. Specifically, in explaining the outcome of a case in question, the disputants put forward as arguments relevant cases favouring their respective positions, with arguments/cases conflicting due to their features, stages and outcomes, and the applicability of arguments/cases arbitrated by the features and stages of the case in question. We in addition use arbitrated dispute trees to identify the excess features that help the winning disputant to win the dispute and thus complement the explanation. We evaluate our novel methodology theoretically, proving desirable properties thereof, and empirically, in the context of primary legislation in the United Kingdom (UK), concerning the passage of Bills that may or may not become laws. High-level factors underpinning a Bill's passage are its content-agnostic features such as type, number of sponsors, ballot order, as well as the UK Parliament's rules of conduct. Given high numbers of proposed legislation (hundreds of Bills a year), it is hard even for legal experts to explain on a large scale why certain Bills pass or not. We show how our methodology can address this problem by automatically providing high-level explanations of why Bills pass or not, based on the given Bills and their content-agnostic features. [ABSTRACT FROM AUTHOR]

Abstract: In this paper we discuss the redesign of a support management system deployed in a small and medium sized enterprise (SME) in the UK. The original system was not fulfilling its needs as it had not captured work practices in a way that was recognizable to the users. The advantages of the redesign included: improved usefulness; improved efficiency and productivity; reduced learning time; improved usability; and increased acceptance among users. The system is used to support complex and distributed cooperative activities taking place in an SME. We evaluated the current system and analysed work practices using a user-centred design and evaluation philosophy. In this paper we discuss how user needs are incorporated into the enhanced design of the support management system. The user-centred design techniques used in this research include interviews, questionnaires, observations and user tests. We present comparative evaluation results that show significant improvement in performance of user tasks using the redesigned support management system. The contribution of this paper is the presentation of a case study to show how a user-centred design and evaluation philosophy can lead to better requirements capture resulting in systems that more accurately capture the users’ conceptual models. [ABSTRACT FROM AUTHOR]

Analyzing the COVID-19 pandemic is a critical factor in developing effective policies to deal with similar challenges in the future. However, many parameters (e.g., the actual number of infected people, the effectiveness of vaccination) are still subject to considerable debate because they are unobservable. To model a pandemic and estimate unobserved parameters, researchers use compartmental models. Most often, in such models, the transition rates are considered as constants, which allows simulating only one epidemiological wave. However, multiple waves have been reported for COVID-19 caused by different strains of the virus. This paper presents an approach based on the reconstruction of real distributions of transition rates using genetic algorithms, which makes it possible to create a model that describes several pandemic peaks. The model is fitted on registered COVID-19 cases in four countries with different pandemic control strategies (Germany, Sweden, UK, and US). Mean absolute percentage error (MAPE) was chosen as the objective function, the MAPE values of 2.168%, 2.096%, 1.208% and 1.703% were achieved for the listed countries, respectively. Simulation results are consistent with the empirical statistics of medical studies, which confirms the quality of the model. In addition to observables such as registered infected, the output of the model contains variables that cannot be measured directly. Among them are the proportion of the population protected by vaccines, the size of the exposed compartment, and the number of unregistered cases of COVID-19. According to the results, at the peak of the pandemic, between 14% (Sweden) and 25% (the UK) of the population were infected. At the same time, the number of unregistered cases exceeds the number of registered cases by 17 and 3.4 times, respectively. The average duration of the vaccine induced immune period is shorter than claimed by vaccine manufacturers, and the effectiveness of vaccination has declined sharply since the appearance of the Delta and Omicron strains. However, on average, vaccination reduces the risk of infection by about 65–70%. [ABSTRACT FROM AUTHOR]

Most existing deep learning based methods for vessel segmentation neglect two important aspects of retinal vessels: The orientation information of vessels and the contextual information of the whole fundus region. In this paper, we propose a robust orientation and context entangled network (OCE-Net), which can extract complex orientation and context information from blood vessels. To achieve complex orientation-aware convolution, we propose a dynamic complex orientation-aware convolution (DCOA Conv) to extract complex vessels with multiple orientations for improving vessel continuity. To simultaneously capture the global context information and emphasize the important local information, we propose a global and local fusion module (GLFM) to simultaneously model the long-range dependency of vessels and give sufficient attention to local thin vessels. A novel orientation and context entangled nonlocal (OCE-NL) module is also proposed to entangle the orientation and context information together. In addition, an unbalanced attention refining module (UARM) is proposed to deal with the unbalanced pixel numbers of the background and thick and thin vessels. Extensive experiments were performed on several commonly used datasets (DRIVE, STARE, and CHASEDB1) and some more challenging datasets (AV-WIDE, UoA-DR, RFMiD, and UK Biobank). The ablation study results show the proposed method's good performance in maintaining the continuity of thin vessels, and the comparative experimental results show OCE-Net's good performance in retinal vessel segmentation. Thus, the proposed framework can effectively carry out retinal vessel segmentation. • An Orientation and Context Entangled Network is proposed for vessel segmentation. • A DCOA convolution is proposed to capture the orientation of vessels. • A Global and Local Fusion Module is proposed to modeling the global dependencies. • An OCE-DNL is proposed to entangle the orientation and context. • An Unbalanced Attention Refine Module is designed to refine the output features. [ABSTRACT FROM AUTHOR]

To boost the application of machine learning (ML) techniques for credit scoring models, the blackbox problem should be addressed. The primary aim of this paper is to propose a measure based on counterfactuals to evaluate the interpretability of a ML credit scoring technique. Counterfactuals assist with understanding the model with regard to the classification decision boundaries and evaluate model robustness. The second contribution is the development of a data perturbation technique to generate a stress scenario. We apply these two proposals to a dataset on UK unsecured personal loans to compare logistic regression and stochastic gradient boosting (SBG). We show that training a blackbox model (SGB) as conditioned on our data perturbation technique can provide insight into model performance under stressed scenarios. The empirical results show that our interpretability measure is able to capture the classification decision boundary, unlike AUC and the classification accuracy widely used in the banking sector. • Measure based on counterfactuals to evaluate global interpretability of models. • Development of a data perturbation technique to generate a stress testing scenario. • Generate different stressed scenarios from a UK unsecured personal loans dataset. • Model performance is negatively impacted when loan default ratio is increased. • Constraints give insight into feature values that correlate with misclassification. [ABSTRACT FROM AUTHOR]

Abstract: Due to globalization, logistics has become an important part in the supply chain. Many logistics service providers have realised the importance of adoption of technologies that can help manufacturers, warehouses, and retailers to communicate with each other more efficiently. Among many logistics technologies, radio frequency identification (RFID) has been identified as an important technology to improve logistics operations and supply chain management, and thus is increasingly gaining both practitioners’ and researchers’ attention. The purpose of this study is to identify the impact of usability features of RFID in the adoption of the technology by the logistics sector in the UK. We have used questionnaire survey method to collect data from the UK logistics firms. The analysis of the data shows that the usability of RFID technology positively influences adoption of technology. We have further tested the moderating effects of firm size, experience with barcode use, and government support in adopting RFID. Our results show that government support strongly moderates the link between usability of RFID and its adoption but size and experience with barcode do not moderate this link. We elaborate the contributions of the study and managerial implications of our results in this paper. [Copyright &y& Elsevier]

Abstract: In the UK alone there are currently over 4.2 million operational CCTV cameras, that is virtually one camera for every 14th person, and this figure is increasing at a fast rate throughout the world (especially after the tragic events of 9/11 and 7/7) (). Security concerns are not the only factor driving the rapid growth of CCTV cameras. Another important reason is the access of hidden knowledge extracted from CCTV footage to be used for effective business decision making, such as store designing, customer services, product marketing, reducing store shrinkage, etc. Events occurring in observed scenes are one of the most important semantic entities that can be extracted from videos (). Most of the work presented in the past is based upon finding frequent event patterns or deals with discovering already known abnormal events. In contrast, in this paper we present a framework to discover unknown anomalous events associated with a frequent sequence of events (A EASP); that is to discover events, which are unlikely to follow a frequent sequence of events. This information can be very useful for discovering unknown abnormal events and can provide early actionable intelligence to redeploy resources to specific areas of view (such as PTZ camera or attention of a CCTV user). Discovery of anomalous events against a sequential pattern can also provide business intelligence for store management in the retail sector. The proposed event mining framework is an extension to our previous research work presented in and also takes the temporal aspect of anomalous events against frequent sequence of events into consideration, that is to discover anomalous events which are true for a specific time interval only and might not be an anomalous events against frequent sequence of events over a whole time spectrum and vice versa. To confront the memory expensive process of searching all the instances of multiple sequential patterns in each data sequence an efficient dynamic sequential pattern search mechanism is introduced. Different experiments are conducted to evaluate the proposed anomalous events against frequent sequence of events mining algorithm’s accuracy and performance. [Copyright &y& Elsevier]

Abstract: As knowledge emerges as the primary strategic resource in the 21st century, many firms in the manufacturing and service sectors alike are beginning to introduce and implement Knowledge Management (KM). Organisations can certainly benefit from its application for enhanced decision support, efficiency and innovation, thus helping them to realise their strategic mission. However, KM is an emerging paradigm, and not many organisations have a clear idea of how to proceed with it. This paper presents the results of a case study conducted in one company in the United Kingdom (UK), the major aim being to identify how it has developed a KM initiative and system. Hopefully, the information extracted from this study will be beneficial to other organisations that are attempting to implement KM or to those that are in the throes of adopting it. [Copyright &y& Elsevier]

• A novel method for investment builds from classic statistics and deep-learning. • Portfolio formation by mean-variance and deep learning for predictive finance. • Emphasis on preselection of high quality assets. • Considering long-term dependences in time-series fluctuation. • The results outperform other strategies with respect to potential returns and risks. Portfolio theory is an important foundation for portfolio management which is a well-studied subject yet not fully conquered territory. This paper proposes a mixed method consisting of long short-term memory networks and mean-variance model for optimal portfolio formation in conjunction with the asset preselection, in which long-term dependences of financial time-series data can be captured. The experiment uses a large volume of sample data from the UK Stock Exchange 100 Index between March 1994 and March 2019. In the first stage, long short-term memory networks are used to forecast the return of assets and select assets with higher potential returns. After comparing the outcomes of the long short-term memory networks against support vector machine, random forest, deep neural networks, and autoregressive integrated moving average model, we discover that long short-term memory networks are appropriate for financial time-series forecasting, to beat the other benchmark models by a very clear margin. In the second stage, based on selected assets with higher returns, the mean-variance model is applied for portfolio optimisation. The validation of this methodology is carried out by comparing the proposed model with the other five baseline strategies, to which the proposed model clearly outperforms others in terms of the cumulative return per year, Sharpe ratio per triennium as well as average return to the risk per month of each triennium. i.e. potential returns and risks. [ABSTRACT FROM AUTHOR]

• Proposed an intelligent diagnosis framework for skin lesion classification. • Designed a multi-class multi-level algorithm to enhance the accuracy. • Proposed improved techniques for noise removal from the images. • Deep learning and other machine learning approaches developed and compared. • So far, the best multi-class result (∼96.5% accuracy) achieved using deep learning. Skin diseases remain a major cause of disability worldwide and contribute approximately 1.79% of the global burden of disease measured in disability-adjusted life years. In the United Kingdom alone, 60% of the population suffer from skin diseases during their lifetime. In this paper, we propose an intelligent digital diagnosis scheme to improve the classification accuracy of multiple diseases. A Multi-Class Multi-Level (MCML) classification algorithm inspired by the "divide and conquer" rule is explored to address the research challenges. The MCML classification algorithm is implemented using traditional machine learning and advanced deep learning approaches. Improved techniques are proposed for noise removal in the traditional machine learning approach. The proposed algorithm is evaluated on 3672 classified images, collected from different sources and the diagnostic accuracy of 96.47% is achieved. To verify the performance of the proposed algorithm, its metrics are compared with the Multi-Class Single-Level classification algorithm which is the main algorithm used in most of the existing literature. The results also indicate that the MCML classification algorithm is capable of enhancing the classification performance of multiple skin lesions. [ABSTRACT FROM AUTHOR]