Centre de Recherche sur l'Information Scientifique et Technique

Finding pertinent articles is extremely difficult for scholars due to the exponential growth of scientific publications. Despite their usefulness, current recommendation systems can fail to catch the semantic and thematic subtleties included in intricate scientific language. To address these limitations, we propose BTG-SR (Bipartite Topic Graph for Scientific Recommendation), a hybrid recommendation system that combines content-based and collaborative filtering approaches. BTG-SR harnesses the power of SciBERT, a pre-trained language model for fine semantic encoding of scientific articles, coupled with BERTopic for state-of-the-art topic modeling. It also incorporates community detection from a bipartite graph of researchers-themes. This combination simultaneously takes into account semantic, thematic, and social dimensions. The experimental results on a multidisciplinary corpus demonstrate that BTG-SR outperforms benchmark methods in terms of precision and recall, thus providing a robust and scalable solution for scientific literature recommendation.

• Optimized concrete mix design using recycled coarse aggregates through DNN and IGWO. • Multi-objective optimization framework reveals trade-offs between density and ultrasonic pulse velocity. • Innovative use of machine learning for modeling water content, density, and UPV in recycled aggregate concrete. • Hybrid approach of Grey Wolf Optimizer and Deep Neural Network achieves superior predictive accuracy. • Validated models offer a sustainable framework for concrete mix design in eco-friendly construction. This study presents a thorough methodology for forecasting and enhancing the essential performance characteristics of concrete with Recycled Coarse Aggregates (RCA), including water content (Wt), density, and Ultrasonic Pulse Velocity (UPV). The attributes are represented as functions of cement content (300–400 kg/m³), RCA percentage (0–100 %), and slump (5–12 ± 1 cm). An experimental database was created using a three-factor, three-level design, followed by exploratory analysis using correlation matrices and three-dimensional response surface plots to identify significant input-output associations. Various modeling techniques, such as Response Surface Methodology (RSM), Decision Tree (DT), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and a Deep Neural Network (DNN) enhanced by the Improved Grey Wolf Optimizer (IGWO), were assessed. The accuracy of the model was meticulously evaluated by 4-fold cross-validation, Taylor diagrams, and radar plots, demonstrating that the DNN-IGWO hybrid surpassed all other models, achieving the minimal prediction errors and the maximal correlation coefficient (R 2 ) values for all target responses. The validated models were later employed as surrogates inside a multi-objective optimization framework employing the Multi-Objective Grey Wolf Optimizer (MOGWO). The optimization produced well-distributed Pareto fronts, emphasizing the intrinsic trade-off between density and UPV. The resultant "knee" solutions identify specific concrete compositions that optimize UPV (3.6–3.7) while ensuring a high density (≥2400 kg/m³) and moderate water content. This study provides a scientifically robust, data-driven paradigm for the sustainable design of concrete mixtures, facilitating the appropriate incorporation of recycled aggregates while enhancing material qualities.

Backgraund. Despite the high prevalence of endometriosis and the social significance of the endometriosis-associated infertility issue, reliable molecular markers of this disease are not known. Members of heterogeneous group of small non-coding RNA appear to be promising markers of the disease, and analysis of these molecules in eutopic endometrium can be a basis of innovative diagnostic technology. Objective. To conduct a comparative analysis of known classes of small non-coding RNAs expression in the material of the eutopic endometrium of patients and healthy women. Material and methods. The study was carried out using «deep» sequencing, the results obtained were verified using microarray PCR — a technology optimal for creating new diagnostic solutions at the next stages of work. Objects of the study — eutopic endometrium samples of patients with external genital endometriosis (n=45) and healthy women (n=24). Results. Potential marker microRNA (microRNA), piRNA (piwi-RNA), small nucleolar RNA (snoRNA) and small nuclear RNA (snRNA) were identified. The difference in the expression activity of a number of microRNAs (miR-143-3p, -320a-3p, -4488, -let-7b-5p, -451a, -16-5p) in the eutopic endometrium of patients with endometriosis vs. healthy women was verified. The high diagnostic potential of the concentration ratio of a number of so-called «miRNA reciprocal pairs» (for example, miR-12136/-320a, AUC = 0.89) was demonstrated. Conclusions. Analysis of short regulatory RNA in eutopic endometrial is a promising PCR-based technology for minimally invasive diagnosis of endometriosis.

The present study aimed to characterize Cystoseira stricta sodium alginate and its maleated derivative for potential use in food preservation. The alginate was extracted from the brown alga Cystoseira stricta , harvested from west Algerian coast (35°49′N/0°01′W). The sodium alginate was grafted with maleic anhydride to obtain maleated sodium alginate. The native and grafted sodium alginate were analyzed using FTIR, XRD, SEM, DSC, TGA, X-ray, and EDX. Their antioxidant potential was evaluated using DPPH scavenging activity. Additionally, a nanoemulsion based on maleated sodium alginate and date seed oil was synthezised and applied as an edible functional coating for delaying strawberries’ spoilage. The extraction of sodium alginate resulted in good yield (37.52%) with a β-D-mannuronic acid /α-L-guluronic acid (M/G) ratio of 0.80. The grafting of sodium alginate with maleic anhydride has been successfully executed and confirmed by infrared analysis and DSC. The degree of substitution was estimated to 93%. The crystallinity becomes greater according to x-ray diffraction data. Thermal analysis indicated that maleated sodium alginate was more thermally stable. The SEM/EDX analyses revealed well defined morphological characteristics and high chemical purity of native and modified sample. The maleated sodium alginate exhibits higher antioxidant activity than native sodium alginate. It was found the maleation cannot improve only the antioxidant ability but also the product esthetic. The prepared nanostructured system demonstrated dispersion stability with a zeta potential of −29.9, as determined by DLS. When applied as an edible coating, it effectively protected and enhanced the shelf-life of fresh strawberries at ambient temperature.

Due to the toxic, carcinogenic, and environmentally persistent nature of dyes such as crystal violet, water pollution resulting from industrial dye discharges has become a critical global environmental issue. This study investigates a hybrid approach to address water pollution by integrating adsorption and advanced oxidation processes (AOPs) using ZnO/TiO2-CaAlg as a catalyst. Sodium alginate, a naturally occurring biopolymer, was employed as the support matrix for the catalyst, taking advantage of its hydrophilic nature, biocompatibility, and ability to form strong electrostatic interactions with cationic dyes. Key parameters such as dye concentration, catalyst dosage, hydrogen peroxide volume, and solution pH were optimized to maximize degradation efficiency. Both individual and combined processes were evaluated to assess the synergistic effects on dye removal. The stability and reusability of the catalyst were evaluated, demonstrating its potential as a scalable and environmentally sustainable solution for wastewater treatment. The findings underscore the potential of ZnO/TiO2-CaAlg catalysts in mitigating persistent organic pollutants, contributing to the development of advanced treatment methods for environmental remediation.

Pharmaceutical contamination of aquatic systems poses an increasing environmental concern due to the persistence, bioactivity, and incomplete removal of these compounds by conventional wastewater treatment processes. This study investigates the adsorptive removal of three structurally distinct pharmaceuticals: ketotifen fumarate (KF), doxycycline hyclate (DXC), and nystatin (Nyst), using raw bentonite (RB). By combining batch experiments with an interpretable machine learning (ML) framework, adsorption kinetics, equilibrium, and thermodynamics were evaluated. Additionally, four Ant Lion Optimizer (ALO)-optimized models: Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost), were employed to predict adsorption capacity under diverse conditions. RB exhibited high adsorption capacities: 178.86 ± 1.26 mg/g for KF, 222.91 ± 2.02 mg/g for DXC, and 190.25 ± 2.86 mg/g for Nyst. Adsorption equilibrium was best described by the Freundlich isotherm, indicating multilayer adsorption on a heterogeneous surface. Thermodynamic and spectroscopic analyses revealed a dual mechanism involving electrostatic attraction, hydrogen bonding, cation exchange, and van der Waals interactions, with DXC and Nyst adsorption being endothermic and KF adsorption exothermic. SHAP (SHapley Additive exPlanations) analysis identified adsorbent dosage, initial concentration, and pH as dominant operational factors, while the molecular descriptor nC (number of carbon atoms) emerged as key to differentiating pharmaceuticals, linking larger molecular size to stronger adsorption. The XGBoost model achieved the highest accuracy (R2 = 0.972, RMSE = 0.1225), demonstrating robust generalizability. These findings highlight RB as a low-cost, scalable adsorbent and establish an interpretable ML approach capable of linking molecular structure to adsorption behavior.