Résumé: This paper presents a simulation system utilizing Long Short-Term Memory (LSTM) forecasting models combined with Monte Carlo methods to predict energy generation and model uncertainty from renewable and fossil fuel sources in the Adrar region of Algeria. The study focuses on short-term load forecasting for the Kabertane wind field, the Adrar solar photovoltaic farm, and overall electrical demand. Separate LSTM models predict each renewable source’s output, which are aggregated and subtracted from load demand forecasts to determine necessary fossil fuel requirements. Monte Carlo methods quantify uncertainty by fitting error distributions to forecast residuals and generating multiple realizations with added noise. This probabilistic approach provides robust assessment of economic costs and environmental impacts of energy production. Results demonstrate significant cost savings and CO2 emission reductions through renewable energy incorporation while emphasizing the critical role of uncertainty modeling in optimizing energy production, cost, and environmental sustainability in the region.

Résumé: The increasing popularity of recommender systems is leading to a focus on improving them using knowledge graphs (KG). This new trend has emerged to address the problem of data sparsity. It is evident that hybrid approaches combining collaborative and content-based filtering have demonstrated considerable potential; however, they do not fully use the semantic richness of knowledge graphs. To address this, we propose a method that enriches the knowledge graph and improves the recommendation performance by identifying implicit relationships between entities using unsupervised clustering and cosine similarity calculations. These relationships are then integrated into the graph in the form of new weighted edges. The TransR model is then used to recalculate the enriched embeddings, before employing an attention mechanism to capture complex dependencies between users and items. Experiments conducted on the Yelp2018 and Amazon datasets demonstrate a substantial enhancement in the efficacy of recommendation systems, underscoring the efficacy of the proposed approach in enhancing attentional propagation and generating recommendations that are more pertinent and tailored to individual users.

Résumé: This paper presents a Multi-Agent System (MAS) approach for designing an air pollution simulator. The aim is to simulate the concentration of air pollutants emitted from sources (eg factories) and to investigate the emergence of cooperation between the emission source managers and the impact this has on air quality. The emission sources are controlled by agents. The agents try to achieve their goals (ie increase production, which has the side effect of raising air pollution) and also cooperate with others agents by altering their emission rate according to the air quality. The agents play an adapted version of the evolutionary N-Person Prisoners' Dilemma game in a non-deterministic environment; they have two decisions: decrease or increase the emission. The rewards/penalties are influenced by the pollutant concentration which is, in turn, determined using climatic parameters. In order to give predictions about the Plume Dispersion) model and an ANN (Artificial Neural Network) prediction model. The prediction is calculated using the dispersal information and real data about climatic parameters (wind speed, humidity, temperature and rainfall). Every agent cooperates with its neighbours that emit the same pollutant, and it learns how to adapt its strategy to gain more reward. When the pollution level exceeds the maximum allowed level, agents are penalised according to their participation. The system has been tested using real data from the region of Annaba (North-East Algeria). It helped to investigate how the regulations enhance the cooperation and may help controlling the air quality. The designed system helps the environmental agencies to …

Résumé: This paper describes the development of a global air quality prediction model based on the combination of five different pollutants predicted values; specifically: O3, PM10, SO2, NOx and COx. Each pollutant concentration prediction is obtained from a radial basis function (RBF) neural network developed in order to predict 12 hours ahead the five air pollutant parameters for the region of Annaba, northeastern Algeria. Given the measurement of air pollutant concentration and three chosen metrological parameters (wind speed, temperature and humidity) at time t, the models can predict the air pollutant concentrations at t+12 hours. Once these concentrations are obtained, a second artificial neural network (ANN) given by a multi-layered perceptron (MLP) is used to combine them and forecast the air quality over a scale ranging from 1 for very good to 5 for very bad.

Résumé: Colposcopy is a medical examination done using an optical tool in order to inspect the lesions in the patient's cervix. It is mainly recommended to screen for cervical cancer. Due to privacy and cultural concerns, images of colposcopy are hard to find, which makes colposcopy training examples very limited. This paper aims to present a Generative Adversarial Network (GAN) based system that anonymizes and generates synthetic images of colposcopy for medical student training. The proposed system can be used by medical teachers and trainers in order to present clinical cases using fake images generated using a GAN. The latter is trained using a dataset of real images of colposcopy labelled by a domain expert. Our current prototype is a web- based application wherein the students can visualise colposcopy images and then have to choose the right diagnostics, which have been previously defined by the teachers. The prototype has shown interesting feedback from the students and the teachers; however, the generated image's quality needs to be enhanced.

Résumé: Shoulder surfing is a common threat used to steal sensitive information, specifically credential and PIN number either by human or recorder cameras. The proposed solution is designed for both PCs and smartphone-based login interfaces. In this paper, we propose a new authentication scheme based on augmented misinformation existing techniques in addition to the introduction of a novel one for maximizing security level and keeping it simple to use. We reuse the crossing-based technique with keys layout randomisation support in addition to displacement factor consideration. Moreover, we introduce white keys-based technique that is used for increasing the security level when keeping the same difficulty-level of exploitation. The evaluation results show that the proposed scheme outperforms well known methods of the state of the art and gives a trade-off between the usability and the security levels against …