This paper looks into the extension of the pymcdm library, focusing on reference point-based techniques. It introduces the implementations of methods such as the Reference Ideal Method (RIM), Preference Ranking On the Basis of Ideal-average Distance (PROBID), and Election based on Relative Value Distances (ERVD). This update is intended to meet the increasing demand for solutions tailored to decision makers’ expertise and experience. Furthermore, it introduces techniques related to sensitivity analysis and weight comparison factors of criteria. By expanding this library, the range of MCDA/MCDM tools is broadened and further progress is encouraged in the use of expert knowledge and the implementation of compromise methods.
This article analyses the performance characteristics of XGBoost models across multiple datasets for phishing URL detection, extending our previous conference paper with comprehensive cross-dataset validation and comparative analysis. Using a validation framework with three distinct datasets — a custom phishing dataset (75,738 samples), the large-scale GramBeddings dataset (639,723 samples), and the PhiUSIIL dataset (47,103 samples)—we demonstrate that XGBoost delivers robust performance across diverse data sources. Our approach addresses the issue of feature instability, showing how balanced feature engineering is crucial for reliable detection. The model achieves 90.7% accuracy and 91.2% F1 score on the custom dataset, with a solid 77.8% average accuracy across three independent test sets. In comparative benchmarks against Neural Networks, XGBoost proved superior in training efficiency and detection quality, achieving 2.1% higher accuracy and training 12.6 times faster (0.95s vs 12.01s) with 3.6 times less memory. Although Neural Networks offered faster inference (7.51ms vs 20.6ms) and smaller model sizes, XGBoost’s balance of high accuracy and rapid training makes it highly effective for practical, real-world phishing detection systems.
Developing efficient ammonia synthesis catalysts is key to reducing energy consumption and improving sustainability. This study explores barium-promoted cobalt catalysts supported on lanthanide oxides (La2O3, Nd2O3, Sm2O3, Eu2O3, Gd2O3) to understand how support choice influences catalytic performance. The catalysts were characterised using techniques such as X-ray powder diffraction (XRPD), high-resolution transmission electron microscopy (HRTEM), and temperature-programmed desorption (H2-TPD, CO2-TPD). Testing under industrially relevant conditions (400-470 degrees C, 6.3 MPa, H2/N2 = 3) revealed that lanthanide oxides strongly affect catalysts' activity, reducibility, and hydrogen adsorption. Among the tested catalysts, the La2O3-supported system exhibited the highest ammonia synthesis activity (ravg = 1.90gNH3 center dot g-cat 1 center dot h-1), likely due to its favorable hydrogen sorption properties and larger active phase surface area available for hydrogen (31 m2 & sdot;gCo-1). These findings highlight the potential of lanthanide oxides as supports and the importance of barium as a promoter in cobalt-based catalysts for ammonia synthesis.
Waste cooking oil (WCO) was investigated as a waste-derived reactive precursor for a cured organic binder in highly mineral-filled composites containing turmeric. Ten formulations selected from a broader experimental screening were prepared from WCO, sulfuric acid, quartz sand, and turmeric added at 1–7% relative to the dry mass of quartz sand. Formulation-specific thermal curing was conducted at 190–210 °C for 12–20 h. Because the binder content, acid-to-binder ratio, turmeric content, curing temperature, and curing time varied simultaneously, the study was designed as an exploratory multifactorial screening rather than as a controlled assessment of individual processing variables. FTIR, 1H NMR analysis of acetone-soluble constituents, TGA, and SEM-EDS were combined with mechanical screening, water-absorption, contact-angle, and microbiological measurements. Bending and splitting tensile strengths ranged from 1.15 to 2.42 MPa and from 0.30 to 0.52 MPa, respectively. Water absorption ranged from approximately 4.8% to 9.0%, while water contact angles exceeded 90° for all investigated formulations. Selected formulations reduced the surviving fraction by more than 90% for Staphylococcus epidermidis and by approximately 70% for Pseudomonas aeruginosa under the applied suspension-assay conditions. The estimated C50 values of 60.6–83.3 mg mL−1 indicated measurable concentration-dependent responses at relatively high nominal composite concentrations. The available results support curing-associated transformation and consolidation of the WCO-derived binder phase but do not quantify crosslink density or retained organic content. Because no matched turmeric-free composite or surface/eluate pH measurements were available, the biological effects are attributed to the complete composite formulations rather than specifically to turmeric or intact curcumin. The results provide an exploratory basis for the further development of waste-derived, non-load-bearing polymer–mineral composites with functional surface and biological properties.
BACKGROUND:In recent years, there has been growing interest in the male reproductive tract microbial contamination and its potential impact on semen quality and the effectiveness of reproductive biotechnologies. OBJECTIVES:The aim of this study was to evaluate the microbial profile of bull semen, determine the antibiotic resistance of isolated bacterial strains, and assess their effect on the expression of SIRT1 as a key protein involved in the cellular response to oxidative stress. METHODS:Semen samples collected during routine cryopreservation procedures were analyzed. Microorganisms were identified, their susceptibility to 10 selected antibiotics was tested, and SIRT1 protein levels were measured at three time points (T0, T24, and T48). Microbial presence was detected in 8 out of 13 semen samples, most commonly Escherichia coli, Streptococcus sp., and Pseudomonas aeruginosa. Significant resistance to antibiotics commonly used in semen extenders, such as penicillin, lincomycin, streptomycin, and tetracyclines, was observed. RESULTS:The highest levels of resistance were found in E. coli and Streptococcus sp. strains. A statistically significant decrease in SIRT1 concentration was also observed in bacterially contaminated samples-from an initial value of 7.09 ng/mL ± 0.53 (T0) to 2.71 ng/mL ± 0.96 after 48 h of incubation (T48). This decline was notably greater than in uncontaminated samples, suggesting an effect of oxidative stress triggered by the presence of pathogens. CONCLUSION:The results of this study indicate a strong association between microbial contamination of semen, oxidative stress, and decreased SIRT1 expression. The growing resistance of bacteria to commonly used antibiotics was also confirmed, posing a challenge to the effectiveness of current sanitary protocols in artificial insemination. SIRT1 may serve as a sensitive biomarker of semen quality and infection presence, providing added value in bull fertility diagnostics.
