Sopra Steria Group

We have developed an 11-question self-assessment test that predicts whether a team is likely to develop accessible digital solutions – or not – based on the characteristics of the development processes. Our results indicate the test can predict both successes and failures with regards to accessibility of digital solutions. As such, teams and product leaders now have an easy way to identify whether the team’s knowledge, practices and mindset makes them likely to deliver accessible digital solutions. Further, the test identify which changes are needed for the team to better ensure digital accessibility.

Ground magnetic field variations have been used to investigate ionospheric dynamics for more than a century. They are usually explained in terms of an electric circuit in the ionosphere driven by an electric field, but this is insufficient to explain how magnetic field disturbances are dynamically established. Here we explain and simulate how the ionosphere dynamically responds to magnetospheric forcing and how it leads to magnetic field deformation via Faraday's law. Our approach underscores the causal relationships, treating the magnetic field and velocity as primary variables (the B, v paradigm), whereas the electric field and current are derived, in contrast to the E, j paradigm commonly used in ionospheric physics. The simulation approach presented here could be used as an alternative to existing circuit‐based numerical models of magnetosphere‐ionosphere coupling.

Generative adversarial networks (GANs) have drawn considerable attention in recent years for their proven capability in generating synthetic data which can be utilized for multiple purposes. While GANs have demonstrated tremendous successes in producing synthetic data samples that replicate the dynamics of the original datasets, the validity of the synthetic data and the underlying privacy concerns represent major challenges which are not sufficiently addressed. In this work, we design a cascaded tabular GAN framework (CasTGAN) for generating realistic tabular data with a specific focus on the validity of the output. In this context, validity refers to the the dependency between features that can be found in the real data, but is typically misrepresented by traditional generative models. Our key idea entails that employing a cascaded architecture in which a dedicated generator samples each feature, the synthetic output becomes more representative of the real data. Our experimental results demonstrate that our model is capable of generating synthetic tabular data that can be used for fitting machine learning models, as CasTGAN’s classification performance only falls under the real training data’s PR-AUC score by 4.88% on average for classification datasets, and exhibits an average reduction of the real training data’s R2 score by 0.139 for regression datasets. In addition, our model captures well the constraints and the correlations between the features of the real data, especially the high dimensional datasets. Assessing the generation of invalid records demonstrates that CasTGAN reduces the number of invalid data observations by up to 622% in comparison to the second best performing baseline tabular GAN model. Furthermore, we evaluate the risk of white-box privacy attacks on our model and subsequently show that applying some perturbations to the auxiliary learners in CasTGAN increases the overall robustness of our model against targeted attacks.

Circular Economy ist ein zunehmend wichtiges Thema in Gesellschaft, Kultur, Politik und Wirtschaft. Die Begrenzung von Rohstoffverschwendung ist für Unternehmen auch im Sinne des Brand Managements von Vorteil, da ein erkennbares Engagement zu mehr Nachhaltigkeit und Umweltschutz zu einer positiveren Wahrnehmung des Unternehmens beitragen kann. Für einen weitreichenden Wandel müssten sich viel mehr Unternehmen an der Kreislaufwirtschaft beteiligen. Dieser Artikel will eine Möglichkeit aufzeigen, wie sich Unternehmen besser austauschen und vernetzen können, um die Kreislaufwirtschaft weiter voranzutreiben. In Zusammenarbeit mit der Unternehmensberatung wg-data, die Firmen unter anderem auch beim nachhaltigen Handeln unterstützt, haben die Autor:innen vorhandene Circular-Economy Plattformen analysiert und auf dieser Grundlage in einem Seminar an der Hochschule Anhalt die Idee einer deutschlandweiten, branchenübergreifenden „Circular Match“-Plattform entwickelt.

Many transportation networks have complex infrastructures (road, rail, airspace, etc.). The quality of service in air transportation depends on weather conditions. Technical failures of the aircraft, bad weather conditions, strike of the company’s staff cause delays and disrupt traffic. How can the robustness of such networks be improved? Improving the robustness of air transportation would reduce the cascading delays between airports and improve the passenger journey. Many studies have been done to find critical links and nodes, but not so many analyze the paths. In this paper, we propose a new method to measure network robustness based on alternative paths. Besides improving the robustness of the French (respectively Turkish Airlines and European) low-cost flight network by 19% (respectively 16% and 6.6%), the method attempts to show the relevance of analyzing the network vulnerability from a path-based approach.

Weather obstacles in the airspace can interfere with an aircraft’s flight plan. Pilots, assisted by air traffic controllers (ATCs), perform avoidance maneuvers that can be optimized. This paper addresses the generation of alternative aircraft trajectories to resolve unexpected events. The authors propose a solution based on the RRG algorithm, K-means clustering, and Dynamic Time Warping (DTW) similarity metric to address the problem. The mixed algorithm succeeds in generating a set of paths with diversity in an obstacle constrained airspace between Paris-Toulouse and London-Toulouse airports. This tool could help to reduce the workload of pilots and ATCs when such a situation arises.

This study introduces a new method for identifying COVID-19 infections using blood test data as part of an anomaly detection problem by combining the kernel principal component analysis (KPCA) and one-class support vector machine (OCSVM). This approach aims to differentiate healthy individuals from those infected with COVID-19 using blood test samples. The KPCA model is used to identify nonlinear patterns in the data, and the OCSVM is used to detect abnormal features. This approach is semi-supervised as it uses unlabeled data during training and only requires data from healthy cases. The method’s performance was tested using two sets of blood test samples from hospitals in Brazil and Italy. Compared to other semi-supervised models, such as KPCA-based isolation forest (iForest), local outlier factor (LOF), elliptical envelope (EE) schemes, independent component analysis (ICA), and PCA-based OCSVM, the proposed KPCA-OSVM approach achieved enhanced discrimination performance for detecting potential COVID-19 infections. For the two COVID-19 blood test datasets that were considered, the proposed approach attained an AUC (area under the receiver operating characteristic curve) of 0.99, indicating a high accuracy level in distinguishing between positive and negative samples based on the test results. The study suggests that this approach is a promising solution for detecting COVID-19 infections without labeled data. Keywords: COVID-19; routine blood tests; kernel PCA; semi-supervised anomaly detection; data-driven

Semiconductor materials provide a compelling platform for quantum technologies (QT). However, identifying promising material hosts among the plethora of candidates is a major challenge. Therefore, we have developed a framework for the automated discovery of semiconductor platforms for QT using material informatics and machine learning methods. Different approaches were implemented to label data for training the supervised machine learning (ML) algorithms logistic regression, decision trees, random forests and gradient boosting. We find that an empirical approach relying exclusively on findings from the literature yields a clear separation between predicted suitable and unsuitable candidates. In contrast to expectations from the literature focusing on band gap and ionic character as important properties for QT compatibility, the ML methods highlight features related to symmetry and crystal structure, including bond length, orientation and radial distribution, as influential when predicting a material as suitable for QT.

This letter presents an effective data-driven anomaly detection scheme for automatically recognizing unbalanced sitting posture in a wheelchair using data from pressure sensors embedded in the wheelchair. Essentially, the designed approach merges the desirable features of the kernel principal component analysis (KPCA) as a feature extractor with the Kantorovich distance (KD)-driven monitoring chart to detect abnormal sitting posture in a wheelchair. It is worth noting that this approach does not require labeled data but only employs normal event data to train the detection model, which makes it more appealing in practice. Specifically, we used the KPCA model to exploit its capacity to reduce the dimensionality of nonlinear data to obtain good detection. At the same time, the KD monitoring scheme is an efficient distribution-driven anomaly detection approach in multivariate data. Furthermore, a nonparametric decision threshold using kernel density estimation is adopted to extend the flexibility of the proposed approach. Publicly available data have been used to verify the detection capacity of the proposed approach. The overall detection system proved promising, outperforming some commonly used monitoring methods.

With the increasing development of artificial intelligence (AI) technologies, deep learning-driven approaches have been widely applied to predicate different machinery failures. One key challenge of failure prediction is to collect sufficient data, especially data of various failure types, to train the data-driven models. Existing studies focus on using transfer learning to transfer knowledge across machines or domains, but not across failure types. In this study, we hypothesise that knowledge about failure among similar failure types is transferable. Should the hypothesis hold, companies may no longer require a large amount of all types of failure data for predictive maintenance. This will increase the companies’ overall implementation feasibility and productivity gains. We tested our hypothesis on knowledge transferability for failure prediction in an experiment performed on rotating machinery with vibration signals. During the experiment, we first calibrated the performance of the trained deep neural network in each impending failure type. Then, we leveraged the architecture and hyperparamesters of the neural network model trained from one type of failure as the pre-trained model for knowledge transfer. The pre-trained model is fine-tuned with data from another type of failure of the same machine. After that, we compared the performance of the neural network model to predict the second type of failure before and after knowledge transfer. Results showed that transferring knowledge obtained from one type of failure could vastly improve the performance of predicting another type of failure, which may not have sufficient data to train a good prediction model. This result implies that predictive analytics can apply parameter-based deep transfer learning (TL) to address the challenge of insufficient data on all types of machine failures for failure prediction.

Following years of intensive international debate of the ethical and human rights implications of artificial intelligence (AI)-related technologies, there are numerous proposals to legislate and regulate these technologies. One aspect of possible legislative frameworks for AI is the creation of a new regulator or other body with the remit to provide oversight of AI. This article reviews the ethical and human rights challenges as well as proposed mitigation strategies, in order to the discuss how a regulatory body might be designed to address these challenges. It focuses on a particular form that a new body might take, more specifically on a potential European Agency for AI. Based on a multi-step methodology of stakeholder interaction, the article proposes a terms of reference for such an Agency and discusses the characteristics it would need to display to ensure that it could adequately engage with current and future ethical and human rights challenges arising from the development, deployment and use of AI. This proposal is then contrasted with the proposed European Artificial Intelligence Board included in the draft European Regulation on AI (the AI Act).

Level lifetimes for the candidate chiral doublet bands of ⁸⁰Br were extracted by means of the Doppler-shift attenuation method. The absolute transition probabilities derived from the lifetimes agree well with the M1 and E2 chiral electromagnetic selection rules, and are well reproduced by the triaxial particle rotor model calculations. Such good agreements among the experimental data, selection rules of chiral doublet bands and theoretical calculations are rare and outstanding in researches of nuclear chirality. Besides odd-odd Cs isotopes, odd-odd Br isotopes in the A≈ 80 mass region represent another territory that exhibits the ideal selection rules expected for chiral doublet bands.

Recently, the hospital systems face a high influx of patients generated by several events, such as seasonal flows or health crises related to epidemics (e.g., COVID’19). Despite the extent of the care demands, hospital establishments, particularly emergency departments (EDs), must admit patients for medical treatments. However, the high patient influx often increases patients’ length of stay (LOS) and leads to overcrowding problems within the EDs. To mitigate this issue, hospital managers need to predict the patient’s LOS, which is an essential indicator for assessing ED overcrowding and the use of the medical resources (allocation, planning, utilization rates). Thus, accurately predicting LOS is necessary to improve ED management. This paper proposes a deep learning-driven approach for predicting the patient LOS in ED using a generative adversarial network (GAN) model. The GAN-driven approach flexibly learns relevant information from linear and nonlinear processes without prior assumptions on data distribution and significantly enhances the prediction accuracy. Furthermore, we classified the predicted patients’ LOS according to time spent at the pediatric emergency department (PED) to further help decision-making and prevent overcrowding. The experiments were conducted on actual data obtained from the PED in Lille regional hospital center, France. The GAN model results were compared with other deep learning models, including deep belief networks, convolutional neural network, stacked auto-encoder, and four machine learning models, namely support vector regression, random forests, adaboost, and decision tree. Results testify that deep learning models are suitable for predicting patient LOS and highlight GAN’s superior performance than the other models.

Digital healthcare platforms have enabled patients to receive healthcare in ways that were impossible previously—for example, by providing a “safer” way to meet, as underscored by the Covid-19 pandemic. This article investigates whether older and younger primary care users display behavioral differences on digital healthcare platforms. The article adopts a mixed-method approach in which one-way ANOVA analysis on a sample of 152 000 patient journeys was combined with qualitative interview data. The findings highlight significant differences in usage between elderly and younger patients. The elderly spends more time during use—for example, during anamnesis, onboarding, and in queues. We also outline how the key antecedent factors that are most central to platform usage, such as perceived usefulness, perceived ease of use, digital maturity, and trust, play out in the elderly user context. The study contributes to the nascent literature on digital healthcare platforms and the postadoption usage of information and communication technologies by the elderly. The article also outlines research implications in the area of DHPs and mHealth for elderly users, and it discusses the practical implications for both platform owners and healthcare professionals, where platform design and information management are particularly important for elderly users.

Forecasting the different types of emergency department (ED) demands (patient flows) in hospital systems much aids ED managers in looking into various options to appropriately allocating the restricted resources available per patient attendance. Deep learning networks have recently gained great success in modeling time-dependent in time series data. Thus, this work advocates the use of deep learning-driven models for patient flows forecasting. Notably, we examine and compare seven deep learning models, Deep Belief Network (DBN), Restricted Boltzmann machines (RBM), Long Short Term Memory (LSTM), Gated recurrent unit (GRU), combined GRU and convolutional neural networks (CNN-GRU), LSTM-CNN, and Generative Adversarial Network based on Recurrent Neural Networks (GAN-RNN), to forecast patient flow in a hospital emergency department. We introduce a forecaster layer as output for each model to enable traffic flow forecasting. Patient flow data from different ED services, including biology, radiology, scanner, and echography, in Lille regional hospital in France, is used as a case study in assessing the considered forecasting models. Four metrics of effectiveness are adopted for evaluating and comparing the forecasting methods. The results show the promising performance of deep learning models for ED patient flow forecasting compared to shallow methods (i.e., ridge regression and support vector regression). In addition, the results highlighted the superior performance of the DBN compared to the other models by achieving an averaged mean absolute percentage error of around 4.097% and R2 of 0.973.

This paper investigates the violations and sanctions that have occurred following the implementation of the General Data Protection Regulation (GDPR). The GDPR came into effect in May 2018 with the aim of strengthening the information privacy of European Union/European Economic Area citizens. Based on existing taxonomies of (i) potential consequences of violating the GDPR (including surveillance, discrimination), (ii) an analysis of 277 sanctions, and (iii) interviews with experts, we offer a mapping of the violations and sanctions almost two years after the regulation was implemented. The most typical complaints were, in descending order: unlawful processing and disclosure of personal information, failure to act on and secure subject rights and personal information, and insufficient cooperation with supervising authorities. Our analysis also indicates an increasing number of fines over time. Regarding size, the fines range from 50,000,000 euros to (symbolic?) 90 euros. While research on GDPR violations and sanctions is somewhat scarce, our study mainly confirms existing findings: that the GDPR is complex and challenging. However, our study provides insight on some of the challenges. Our contribution is mainly practical and aimed at managers in any organization whose goal is to protect information privacy and to learn from the mistakes made by other companies. We also welcome more research on the topic.