École des Sciences de I'information

Meliho M., Boulmane M., Khattabi A., Dansou C.E., Orlando C.A., Mhammdi N., Noumonvi K.D.

Soil organic carbon (SOC) is an essential component, which soil quality depends on. Thus, understanding the spatial distribution and controlling factors of SOC is paramount to achieving sustainable soil management. In this study, SOC prediction for the Ourika watershed in Morocco was done using four machine learning (ML) algorithms: Cubist, random forest (RF), support vector machine (SVM), and gradient boosting machine (GBM). A total of 420 soil samples were collected at three different depths (0–10 cm, 10–20 cm, and 20–30 cm) from which SOC concentration and bulk density (BD) were measured, and consequently SOC stock (SOCS) was determined. Modeling data included 88 variables incorporating environmental covariates, including soil properties, climate, topography, and remote sensing variables used as predictors. The results showed that RF (R2 = 0.79, RMSE = 1.2%) and Cubist (R2 = 0.77, RMSE = 1.2%) were the most accurate models for predicting SOC, while none of the models were satisfactory in predicting BD across the watershed. As with SOC, Cubist (R2 = 0.86, RMSE = 11.62 t/ha) and RF (R2 = 0.79, RMSE = 13.26 t/ha) exhibited the highest predictive power for SOCS. Land use/land cover (LU/LC) was the most critical factor in predicting SOC and SOCS, followed by soil properties and bioclimatic variables. Both combinations of bioclimatic–topographic variables and soil properties–remote sensing variables were shown to improve prediction performance. Our findings show that ML algorithms can be a viable tool for spatial modeling of SOC in mountainous Mediterranean regions, such as the study area.

Meliho M., Khattabi A., Zejli D., Orlando C.A., Dansou C.E.

Housni H., Amrous N., Daoudi N., Malzi M.J.

Morocco's exploration of nuclear energy aligns with both climate goals and national energy ambitions, offering a promising low-carbon alternative to conventional fossil fuels. Despite the nation's significant strides in renewable energy, nuclear power remains understudied, revealing a critical literature gap. This research underscores the global imperative to transition to net-zero emissions and the pivotal role nuclear energy plays in addressing climate change. Within the context of Morocco's 2030 plan, which prioritizes renewable energy, nuclear energy stands as an underexplored aspect, lacking comprehensive research. To bridge this gap, the study employs a PESTLE analysis to examine the political, economic, societal, technological, legal, and environmental factors influencing Morocco's nuclear energy landscape. The integration of insights from various sources, including press releases, reports, and scientific publications, ensures a holistic and well-informed perspective on Morocco's nuclear industry. The paper concludes by providing an overview of nuclear energy use on different scales, accompanied by a detailed discussion of the PESTLE analysis outcomes. This approach seeks to contribute valuable insights for informed decision-making and strategic planning in the realm of nuclear energy development.

Rahimi H., Mezrioui A., Daoudi N.

Trust is a decisive factor in e-services and especially in e-commerce. E-customers usually rely on others’ opinions, reviews, recommendations on products, and services to make the right purchase decision. Nevertheless, deceptive reviewers deliberately disseminate fake and dishonest reviews to falsify the products’ reputation. Consequently, there is a need for Trust and Reputation Assessment to aggregate these text reviews and compute their related reputation scores. For this purpose, Natural Language Processing cannot be omitted from the process of generating reputation scores. In this paper, we propose a Trust and Reputation System named SentiTrustCom STC which is composed of two subsystems: (1) A Combined Idiomatic Ontology-based Sentiment Orientation System that employs NLP techniques and extends SentiWordNet to analyze Text reviews and compute their related Sentiment orientation scores; (2) Trust and Reputation Engine that proposes algorithms to generate reliable Trust and Reputation scores using the generated Sentiment Polarities as inputs. STC aims to analyze the users’ behavioral intention in order to detect any ill-intentioned interventions that could falsify the products’ reputation and hence distort the overall trust among reviewers.

Abdallaoui Maan N.

The study aimed at exploring reading acquisition in Arabic from the perspective of a newly-literate adult native speaker, seeking to improve her reading ability and recite the Quran. Drawing mainly...

Ed-Daoudi R., Alaoui A., Ettaki B., Zerouaoui J.

Precision agriculture techniques have been increasingly adopted worldwide to optimize cultivation practices and achieve sustainable crop production. In this study, we developed a Machine Learning approach to identify optimal cultivation practices for sustainable apple production in precision agriculture in the Msemrir town Morocco. We collected a dataset of cultivation practices and apple yield and size data from 10 farms in the town and used correlation-based feature selection and three Machine Learning algorithms (Linear Regression, Decision Tree, and Random Forest) to develop predictive models. The results showed that irrigation, fertilization, and pruning are the most important cultivation practices for apple production in the region, and the Random Forest model performed the best in predicting apple yield and size based on the selected practices. The use of Machine Learning techniques can help farmers optimize cultivation practices and achieve sustainable apple production by reducing inputs such as water and fertilizer and minimizing environmental impact. Moreover, the use of precision agriculture techniques can help farmers meet consumer demand for sustainable and high-quality apple products.