Ecole Supérieure de Commerce de Tunis

This study explores a method for detecting and isolating faults in wind power conversion systems that use a permanent magnet synchronous generator (PMSG). The goal is to enhance the reliability of the inverter by implementing an fault detection method and fault-tolerant control. The converter incorporates an extra leg with two switches that can be quickly connected in place of a malfunctioning leg, ensuring continued operation. Our fault detection and isolation module employs a threshold-based approach, utilizing real-time signal measurements to identify anomalies exceeding predefined limits. The experimental validation of this module is conducted on a high-performance dSpace DS1104 controller board using MATLAB Simulink. Finally, a real-world experiment was conducted to confirm the effectiveness of the proposed fault detection and isolation (FDI) algorithm. The results successfully demonstrated the algorithm’s ability to detect and isolate switch faults.

The significance of green finance (GF) has grown as its ideas and practices have advanced and matured. It now plays a crucial role as a financial catalyst for promoting sustainable development and reducing carbon emissions. However, the imbalance of green finance clustering (GFC) at the geospatial level in China has become more and more prominent, and there is a lack of research on the impact of GFC on total factor carbon productivity (TFCP). Therefore, the objective of this study is to address this research gap by using panel data from 30 provinces in China during the period of 2007–2018 in order to investigate the impact of the GFC on TFCP. The results of the analysis demonstrate a significant enhancement of GFC on TFCP. Secondly, our basic conclusion still holds after a series of robustness tests and addressing endogenous issues. Thirdly, the positive effect of green security clustering on TFCP is most obvious, followed by green investment clustering and green credit clustering. At last, the impact of GFC on TFCP exhibits nonlinear characteristics, influenced by factors such as energy efficiency, green innovation, and energy consumption. Building upon these findings, this paper puts forth policy recommendations aimed at elevating the GFC level to enhance TFCP, thereby facilitating the achievement of carbon neutrality.

This work investigates the complex dynamics of digitally transforming complex public healthcare organizations. It focuses on the interaction between people, processes, and systems as it relates to the digital transformation of a procurement department in a Middle Eastern public healthcare institution. Based on more than fifty interviews with stakeholders—including end users and procurement staff—as well as an examination of corporate records and procurement procedures, the study emphasizes the negative consequences of introducing digital systems without first updating operational procedures and organizational frameworks. Significant differences between the new digital system and the current processes are found in the findings, which result in inefficiencies, a rise in mistakes, and a decline in staff morale. The study emphasizes the value of a comprehensive strategy for digital transformation that synchronizes people, technology, and processes. It also highlights the necessity of change management techniques and strategic integration. This case study contributes to change management theory by demonstrating the limitations of a technology-centric approach and highlights the necessity of considering the three pillars of transformation for successful digital initiatives.

Advancements in tracking technologies like GPS, RFID and mobile devices have made trajectory data collection widespread. This surge in tracking device usage and location-based services popularity has greatly increased moving object trajectory data availability. The ontological modelling of this kind of data is of paramount importance in understanding and utilising such data effectively. By incorporating maximum semantic data into this model, a variety of essential elements related to mobile object trajectories can be captured. An ontology model rich in semantics not only accurately represents trajectory characteristics but also links them to other relevant elements such as spatial and temporal contexts, movement types and mobile object behaviours. This semantic richness grants the model great adaptability, allowing it to be reused in various contexts related to object mobility and making it generic. Moreover, by integrating this semantic data, the process of analysis and decision-making experiences significant improvement, as it relies on more comprehensive and well-structured information, thereby facilitating informed conclusions and effective strategy implementation. Our objective is to propose a generic ontological model for trajectory data that is rich in semantics and considers the various aspects of moving objects, their movements, their trajectories and their interactions with their environment, aiming to fill the gap identified in other models proposed in the literature.

The purpose of this paper is to examine the impact of financial inclusion on corporate investment. More specifically, this paper investigates the possibility of a nonlinear relationship between financial inclusion and corporate investment, as well as the moderating effect of audit quality in this this relationship. To do so, we selected a group of 400 listed non-financial firms in the Middle East and North Africa (MENA) region (Egypt, Jordan, Kuwait, Morocco, Oman, Qatar, Saudi Arabia, Tunisia, and the United Arab Emirates) over a period from 2007 to 2020. The results show an inverted U-shaped link between financial inclusion and corporate investment, applying the system generalized method of moments (SGMM) method. They also found that audit quality is identified as a moderating factor in the relationship between financial inclusion and corporate investment. Our results show that the interaction of financial inclusion and audit quality improves investment efficiency, but that underinvestment scenario could result from spending free cash-flows on risky projects. The findings of this study could be a valuable contribution to the development of financial inclusion policies and to improving access to credit for policymakers and managers in the MENA region. The combination of financial inclusion and audit quality (internal and external) is indispensable for reducing agency costs and optimizing financial inclusion levels.

Human Activity Recognition plays a crucial role invarious applications extending from healthcare to smart envi-ronments. In this paper, we present a novel approach for HARusing edge computing resources. Our method combines spectralanalysis with deep learning techniques to efficiently extractrelevant features from raw sensor data. By employing spectralanalysis, we successfully reduce the feature space from 561 to39, while maintaining high accuracy, low loss, and improvedperformance metrics such as F1 score and recall. Furthermore,we demonstrate the adaptability of our approach by deployinga quantized neural network model onto a resource-constrainededge device, specifically the NodeMCU microcontroller. Thisenables real-time HAR inference at the edge, making our solutionsuitable for applications where computational resources arelimited. Experimental results on the UCU HAR dataset validatethe effectiveness and efficiency of our proposed method for edge-based human activity recognition.

A sustainable city is a smart city with a minimal impact on the environment, by incorporating technologies to reduce pollution. Traffic congestion which is a major concern contributes to global warming and climate change. Traffic forecasting projects future traffic patterns, using historical and current data to enhance traffic flow management. We propose a whole novel approach for predicting traffic congestion rate based on air quality data. We developed a new ensemble voting model based on Long Short Term Memory (LSTM) and Polynomial Regression (PolReg) models that use a new voting thresholded algorithm instead of the existing voting ones. The hyperparameters were optimized with the Genetic Agorithm, to overcome the non-stationarity of time series. A comparative study with the literature confirmed that our framework outperforms existing researches by keeping an absolute effectiveness according to learning curves, with Mean Absolute Error of 0.04, R-Squared of 0.93, and Root Mean Square Error (RMSE) of 0.05.

This research study investigates the impact of cryptocurrency on economic growth in 10 countries across Asia during past period 2013–2020. For this purpose, the sample included a range of economies with the highest number of Bitcoin usages or transactions. For this purpose, the sample included a range of economies with the highest number of Bitcoin usages or transactions, according to recent international rankings. With specific empirical means on panel data modelling, we attempted to show that economic growth, was influenced negatively by the cryptocurrency ‘Bitcoin’, and we concluded that this instrument leads to an increase in the inflation rate in a country according to the quantity theory of money and leads to a disorder in the monetary policy of a country. Our framework shows that the economic growth proxy was substantially influenced positively by economic indicators such as technology, investment and education but negatively by the high rate of participation, which caused an increase in unemployment. Our empirical results offer insights and insist on the importance of the intervention of the authorities to oversee and control the use of the cryptocurrency Bitcoin to avoid its negative effects and implement a strategy that overcomes these effects from a macroeconomic perspective. JEL Classification: E22, E24, E42, F43, E44, E52

The news of COVID-19 that has been circulating on social media has caused melancholy, anxiety, and terror. One of these social networks is Twitter, that can be considered as a microblogging social networking that allows users to communicate via tweets up to 140 characters. Relevant hashtags, Uniform Resource Locators (URL)s, images, emojis, animated gifs, and videos can all be included in the tweets. By analyzing Covid19 hashtags on Twitter, it is possible to detect the public’s feelings and opinions about this crisis. Understanding the thoughts concealed behind a word or a phrase can help better comprehend the feelings shared on social media. and can be accomplished through Sentiment Analysis (SA). This study aims to investigate the usefulness of hashtag Covid19 tweets in identifying the emotional polarity of the COVID-19 pandemic on Twitter via a SA process. This research’s methodology produced a revolutionary Ensemble classifier, in which a boosting approach called Random Forest (RF) is combined with Bidirectional Encoder Representations from Transformers (BERT) in the same model called Ens-RF-BERT. The sentiment ratings are classified as good, negative, or neutral using an ensemble voting model based on (RF) and the (BERT). A dataset that contains tweets with the hashtag Covid19 was utilized in this study for the categorization, pre-processing, and exploratory data analysis of the tweets. The experimental assessment verifies that, when compared to individual Machine or Deep Learning classifiers used in the literature, the suggested Ensemble Voting Machine and Deep Learning approach (Ens-RF-BERT) classifier outperformed all Machine Learning approaches with 94 percent of the findings. Ens-RF-BERT performs better on the same dataset with an accuracy of 93.01 percent, precision of 94.03 percent, and recall of 93.05 percent.

The emerging trend of Health Information Retrieval (HIR) aims to efficiently address users’ specific information requirements. However, a beyond challenge arises in the healthcare domain due to the necessity for specialized dictionaries, which influence the outcomes of HIR. The Vocabulary Mismatch (VM) phenomenon necessitates more robust efforts in the Health Information Retrieval domain, mainly in query formulation. It is crucial to support clinicians, biomedical scientists, and non-specialists in their daily retrieval endeavors. In this paper, we propose an innovative approach that combines deep image captioning for clinical diagnosis generation then MetaMap normalization, and finally a user involvement step. This combination aims to optimize query formulation task for an enhanced query and an efficient HIR process. Experimental results, conducted on widely used search engines such as Google and Bing, reveal that our approach has demonstrated its effectiveness by enhancing result quality and delivering documents from reliable sources. It has significantly improved the user experience, ensuring relevant results appear within the top five ranking links. The findings demonstrate that the integration of various techniques is a valuable enhancement to the Query Expansion (QE) process, resulting in a notable increase in weighted precision rates by around the twofold a MAP to over 70% and an apparent reduction in Vocabulary Mismatch for Healthcare Information Retrieval.

We investigate theoretically and empirically the effect of globalisation on informality. We use a model with two identical countries. Firms in each country choose to be formal or informal. Relative to an informal firm, a formal one produces a good of higher quality and pays social contributions on each worker. We determine the effect of globalisation on the size and the share of informality from the comparison of Autarky and Full Integration scenarios. We prove that for a given quality gap between formal and informal products, (i) globalisation increases the size of informality for low and high levels of social contributions but decreases it for intermediate ones; (ii) globalisation increases the share of informality for low social contributions and decreases it for high social contributions. The turning points depend increasingly on the quality gap. We then test these theoretical results by relying on the well‐informed data available for Latin‐American countries and using the economic sector as a proxy for the quality gap between a formal and an informal firm. Our empirical results are highly consistent with the theoretical model. In terms of policy implications, they show that globalisation may be used by the states together with modulated social contributions so as to reduce informality, but not for all economic sectors simultaneously.

Epileptic seizure prediction using multi-channel electroencephalogram (EEG) signals is very important in clinical therapy. A large number of channels lead to high computational complexity with low model performance. To improve the performance and reduce the overfitting that arises due to the use of unrelevant channels, the present paper proposed a channel selection method to study the brain region activation related to epileptic seizure. Our method is bio-inspired and cognitive since it integrates the novel binary many-objective particle swarm optimization with a ConvLSTM model. The proposed method has two advantages. First, it performed a new initialization strategy based on channel weighting with mutual information, thereby promoting the fast convergence of the optimization algorithm. Second, it captures spatio-temporal information from raw EEG segments thanks to the ConvLSTM model. The selected sub-channels are optimized as many-objective optimization problem that includes maximizing F1-score, sensitivity, specificity, and minimizing the ratio rate of selected channels. Our results have shown a performance of up to 97.94%$97.94\%$ with only one EEG channel. Interestingly, when using all the EEG channels available, lower performance was achieved compared to the case when EEG channels were selected by our approach. This study revealed that it is possible to predict epileptic seizures using a few channels, which provides evidence for the future development of portable EEG seizure prediction devices.

The aim of this study was to examine the curvilinear relationship between tangible investment and sustainable firm growth in the MENA region, as well as the moderating role of financial inclusion on this connection. To achieve this, we selected a sample of 465 firms over the period 2007–2020. Employing a system GMM model for the empirical analysis, the findings reveal that there is a curvilinear (inverted U-shaped) nexus between tangible investment and sustainable firm growth. Moreover, this study employs a moderating effect model to demonstrate that financial inclusion can enhance sustainable firm growth. The system GMM model further indicates that financial inclusion moderates the curvilinear relationship between tangible investment and sustainable firm growth. This study offers valuable insights for strategic firm planning and policy development, highlighting the role of financial inclusion in promoting firm sustainability.

In this paper, we propose a novel codimension-three-parameter bifurcation analysis of equilibria and limit cycles when integrating Renewable Energy Sources (RESs) power plants with an exponential static load model. The study investigates the effect of solar photovoltaic generation margin, wind power generation margin, and loading factor on the local bifurcation of the modified IEEE nine-bus system. The proposed technique considers the real case of the West System Coordination Council (WSCC), the western states of the USA, by using specific models of RES power plants and static loads. The proposed technique helps to create a set of linearly varying parameters for critical operating points of nonlinear systems. The study explores detailed voltage stability analysis through the examination of bifurcation diagrams. The Hopf, limit-induced, and saddle-node bifurcation branches are identified, defining the parameter space’s stable and unstable operational regions. Additionally, the stability regions surrounding the equilibrium points are diligently explored, clarifying the consequences that various bifurcations may exert on these regions. The study offered in this proposed work aids in determining the best ways to monitor and improve these margins while considering system variables and load design.