Université Pédagogique Nationale

Managing continuous inventory represents a major challenge for commercial enterprises, as it requires ensuring product availability while controlling the costs associated with stockouts and overstocking. Traditional approaches, such as the Wilson model (EOQ), allow the determination of the economic order quantity; however, they remain limited when dealing with dynamic demand fluctuations. This study proposes a hybrid approach that combines the principles of mathematical inventory management with supervised machine learning models, including Random Forest, Support Vector Machine, and Gradient Boosting, integrated within a stacking framework to improve demand forecasting.The experiments are conducted using a dataset extracted from an inventory management system of a large commercial enterprise available on the Kaggle platform. The dataset includes information related to products, sales transactions, orders, dates, and replenishment statuses. The performance of the models is evaluated using statistical metrics such as the coefficient of determination (R²), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE), enabling an objective assessment of forecasting accuracy. The results demonstrate that the combination of classifiers significantly improves prediction accuracy compared with individual models. The developed models are integrated into a responsive web-based business intelligence application that provides inventory managers with real-time visualization of key indicators, including products at risk of stockouts, optimal order quantities, and variations in average stock levels. This decision-support interface facilitates inventory monitoring and enables more proactive supply planning. Overall, the results indicate that the proposed approach improves the reliability of demand forecasts and supports more efficient replenishment decisions while reducing operational costs associated with stock shortages and excess inventory.

<p>Les observations faites avec des lunettes scientifiques font lire que les écoles privées de Kinshasa fixent les frais scolaires en monnaie étrangère. Par voie de conséquence, les parents d’élèves doivent procéder par la conversion de la monnaie locale en monnaie étrangère afin d’effectuer le paiement de ces frais dans le but d’assurer la scolarité de leurs enfants. Pour valider cette hypothèse, nous avons à travers cette étude, analysé les liens existants entre les fluctuations du taux de change et la scolarité des enfants consommateurs des services éducatifs du marché scolaire privé de Kinshasa. Cette investigation a abouti aux résultats selon lesquels : (i) il y a des liens entre taux de change et scolarité, (ii) les frais scolaires sont fixés en monnaie étrangère alors que les parents d’élèves sont rémunérés en monnaie locale, (iii) la hausse du taux de change est un facteur explicatif du décrochage scolaire et de l’inefficacité interne du système éducatif, (iv) les inscriptions des enfants dans les écoles privées moyennant une contribution financière symbolique sont tributaires à la hausse du taux de change.</p><p>Scientific observations indicate that private schools in Kinshasa set tuition fees in foreign currency. Consequently, parents are required to convert local currency into foreign denominations to settle these fees and ensure their children's education. To validate this hypothesis, this study analyzes the correlation between exchange rate fluctuations and the schooling of children within Kinshasa's private education market. The investigation yielded the following results: (i) a significant link exists between exchange rates and schooling stability; (ii) tuition fees are indexed to foreign currency while parents' income is denominated in local currency; (iii) rising exchange rates serve as an explanatory factor for school dropout rates and the internal inefficiency of the educational system; (iv) the enrollment of children in low-cost private schools is increasingly dependent on exchange rate hikes.</p><p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/soc/0807/a.php" alt="Hit counter" /></p>

La présente étude analyse l’influence de l’hétérogénéité des substrats sur la structure des peuplements de macroinvertébrés benthiques dans la rivière Lukunga, un cours d’eau urbain tropical de la ville de Kinshasa (RDC), durant la saison sèche de l’année 2024. Les paramètres physico-chimiques de l’eau, la composition et la distribution des substrats, ainsi que la répartition des macroinvertébrés benthiques ont été étudiés dans cinq stations. Les résultats montrent une relative homogénéité spatiale des paramètres physico-chimiques, sans différence significative entre les stations (test de Kruskal-Wallis, p > 0,05). En revanche, la distribution des macroinvertébrés benthiques varie significativement selon les types de substrats dans plusieurs stations (p < 0,05), traduisant une structuration des communautés à l’échelle du micro-habitat. Les substrats complexes associant sable, vase, matière organique et végétation émergente abritent des peuplements plus riches et plus abondants que les substrats simples dominés par le sable. Les analyses de similarité, ainsi que les analyses multivariées (ACP et CAH), confirment que la composition des communautés benthiques est principalement contrôlée par l’hétérogénéité des substrats plutôt que par les gradients physico-chimiques longitudinaux. Ces résultats soulignent l’importance de la diversité des habitats benthiques dans le fonctionnement écologique des rivières urbaines tropicales et mettent en évidence la pertinence des macroinvertébrés benthiques comme indicateurs de la qualité des habitats.

L'asthme représente un défi de santé publique dans de nombreuses communautés en RDC, incitant les populations à se fier au savoir-faire de la médecine traditionnelle. Cette étude a pour objectif de documenter le patrimoine végétal utilisé localement contre l'asthme dans le secteur Kipuku. Une enquête ethnobotanique a été menée sur le terrain en 2024 auprès de 50 phytothérapeutes 33 hommes (66%) et 17 femmes (34%). Les données recueillies révèlent un savoir traditionnel conséquent, principalement détenu par des individus ayant atteint le niveau secondaire 26 personnes (52%) de l'échantillon.L’analyse a permis de répertorier 36 espèces médicinales distinctes, réparties dans 26 familles botaniques et 16 ordres. Les familles les plus représentées en termes d'espèces antiasthmatiques sont les Euphorbiaceae (16,7%) et les Fabaceae (11,1%). La macération est le mode de préparation dominant (50%). La feuille est l'organe végétal le plus utilisé (23 sujets soit 63,9%), et la voie orale représente la principale voie d'administration (40 sujets soit 80%) des remèdes. Ce savoir traditionnel est majoritairement transmis par voie orale.Ces résultats témoignent d'une richesse floristique et d'un potentiel thérapeutique significatif dans la pharmacopée traditionnelle de Kipuku. Ils constituent une base de données essentielle pour des recherches ultérieures en phytochimie visant l'identification et l'isolement des métabolites secondaires naturels (antiasthmatiques) prometteurs pour le développement de nouvelles solutions thérapeutiques traditionnelles améliorées.

Wood density is a key plant property, indispensable for estimating forest biomass. Yet, despite tropical regions' substantial contributions to global tree diversity and carbon cycling, they remain underrepresented in wood density datasets such as the CIRAD and Global Wood Density Database (GWDD). To address this gap, we present the 'Tervuren xylarium Wood Density Database' (TWDD), containing 13,332 samples from 2,994 species, 1,022 genera, and 156 plant families across six continents (72% from Africa). TWDD offers direct measurements of oven-dry (oven-dry mass/oven-dry volume, all samples), air-dry (air-dry mass/air-dry volume, 6,408 samples), green (green mass/green volume, 1,657 samples), and basic wood density (oven-dry mass/green volume, 1,686 samples). Basic density was estimated for the remaining 11,646 samples via conversion from oven-dry density. TWDD closes a substantial wood density data gap, especially in Africa, adding 1,164 new species, 160 new genera, and 8 new plant families not included in GWDD or CIRAD datasets. The TWDD provides a critical resource for advancing research on forest community dynamics, ecosystem functioning, carbon cycling, and trait-based ecology worldwide.

This study explores the application of gradient descent optimization and loss functions for depression prediction through multidimensional analysis of behavioral, psychological, and social factors. A logistic regression model was trained on a set of 1,631 records. Experimental results indicate an overall accuracy of 91.17 %, a weighted precision of 0.92, a recall of 0.91, and an F1 score of 0.91. These performances demonstrate the robustness of the model and its potential for early detection of depression. This research highlights the role of artificial intelligence in decision support and prevention in mental health. This article explores in depth the gradient descent algorithm, a fundamental pillar of machine learning, focusing on its essential interaction with the loss function (or cost function). Far from a simple theoretical analysis, this talk applies these fundamental concepts to a crucial public health problem: the prediction of depression. Depression is a complex disorder, and its diagnosis and early prediction remain major challenges. To develop reliable predictive models of depression, it is imperative to identify and integrate a diverse set of relevant factors. We will rely on a set of 24 input variables, grouped into 8 symptomatic and contextual categories (such as concentration difficulties, feelings of guilt, sleep disturbances, intense fatigue, etc.), to build our predictive model. The challenge is to navigate the space of these multidimensional parameters in order to find the optimal combination minimizing the prediction error of depression. This paper details how gradient descent, guided by a carefully chosen loss function, addresses this prediction challenge: Modeling of input variables, Application and optimization of gradient descent and Convergence and performance analysis. By integrating these specific variables into a rigorous gradient descent optimization approach, this paper will seek to demonstrate the feasibility and effectiveness of a machine learning model for more accurate and meaningful prediction of depression.

This study, titled “Determinants of Household Poverty in the City-Province of Kinshasa,” examines the key factors influencing household poverty to inform public policy. The objective was to identify the socio-economic characteristics and the physical and financial assets that shape the likelihood of being poor. The analysis is based on a survey of 300 households from Gombe, Bandalungwa, and Kinseso, complemented by data from the 2022–2023 EICVKM. The FGT indices show that 74% of households live below the poverty line, with a poverty depth of 0.358 and severity of 0.191. The ordered logit model reveals that higher education and household income significantly reduce poverty risk. Access to safe water, electricity, decent housing, agricultural inputs, and credit also lowers this probability. The findings highlight the crucial role of human, physical, and financial capital in improving household welfare. The study recommends strengthening education, social infrastructure, financial inclusion, and small economic activities to sustainably reduce urban poverty in Kinshasa.

Studie reaguje na terminologickou nejednoznačnost ve výzkumu didaktiky českého jazyka ve vztahu k žákovu (před)porozumění. Na základě analýzy domácí i zahraniční literatury a s oporou o Rickertovo rozlišení paritního (analytického) a polaritního (hodnotícího) rozlišování vymezujeme vztahy mezi pojmy představa, prekoncept a koncept. Ukazujeme, že část dosavadní terminologie směšuje popisné a hodnotící dimenze, a argumentujeme pro udržení triády představa – prekoncept – koncept jako analytických (paritních) kategorií. Jako kurikulární ekvivalent, který umožňuje explicitní hodnotové (polaritní) rozlišování vzhledem k didakticky transformovanému vědeckému obsahu, navrhujeme užívání termínu žákovo pojetí učiva. Článek tak má ambice přispět k diskusi o vymezení adekvátní terminologie ve výzkumech didaktiky českého jazyka. Význam článku vidíme v potenciálu přispět 1. ke kultivaci komunikace výsledků mezi výzkumníky a učiteli, 2. ke zpřesnění interpretace dat získaných ve výzkumech efektů výuky.

Africa witnessed several reformers and revivalists in Islam before the advent of the European colonists, who succeeded in the partition of Africa. Among them was Sheikh ‘Uthman Dan Fodio, who is supposed by some historians and scholars in civilization to be the most prominent revivalist since he managed to revive Islam in Hausaland and was able, with his followers, to found the Sokoto Caliphate. The present study deals with this eminent revivalist in Islam in Hausaland. The paper begins with a brief biography of this reformer. Then, it explores some aspects of Islamic revivalism made by this outstanding figure. Last but not least, it seeks to reveal whether his revivalist aspects were a success or a failure. This research is considered to be significant since it deals with some aspects of Islamic Revivalism carried by the Shehu ‘Uthman Dan Fodio during his era which was characterized by mingling heretical practices with Islamic beliefs. Thanks to his deep knowledge in Islamic Sciences and his high skills as well as his good morals, he managed to carry out some aspects of Islamic Revivalism mainly those related to women.

This study examined the levels of trace elements in Brassica pekinensis Lour. collected from urban gardening sites (Cecomaf and Lutendele) in Kinshasa (Democratic Republic of the Congo) and associated potential human health risks for the consumer. The toxic metal concentrations in leaf samples were analysed using Inductively Coupled Plasma Mass Spectrometry (ICP-MS) and Atomic Absorption Spectrophotometer (AAS). Results showed that Cr, Cu, Cd, Pb, and Hg concentrations in samples from both sites exceeded the FAO/WHO maximum permissible limits, with Cecomaf displaying notably high levels of Cu (115.39 mg·kg⁻¹), Cd (2.95 mg·kg⁻¹), Pb (30.94 mg·kg⁻¹), and Hg (0.058 mg·kg⁻¹). Health risk assessments associated with the consumption of Brassica pekinensis L, based on Estimated Daily Intake (EDI), Target Hazard Quotient (THQ), and Hazard Index (HI), indices indicated that EDI values for Cd and Pb exceeded recommended limits in several samples, while THQ values exceeded 1 for most metals except As, suggesting potential non-carcinogenic risks. HI values ranged from 2.98 to 16.11 at Cecomaf and 4.85–12.15 at Lutendele, highlighting a high cumulative risk from individual and multi-metal exposure to consumers. Monte Carlo simulation further confirmed these findings by showing that a significant proportion of the exposed population exceeded the WHO safety thresholds for Cd, Pb, and Cu, thereby reinforcing the robustness of the deterministic risk estimates and illustrating the uncertainty associated with age, sex, and local consumption variability. Spearman correlation analysis revealed strong positive associations between Cr–As, Cr–Pb, and Cd–Hg, suggesting common anthropogenic sources or shared environmental pathways. Results of this study highlight the importance of continuous monitoring of trace elements in leafy vegetables for food safety evaluation, adopting stricter agricultural practices in policy making and management, and public health measures to mitigate long-term exposure risks to urban and peri-urban areas populations.