New York Times

The high prevalence and associated negative consequences of intimate partner violence (IPV) (23–49%) against girls has been well-documented in Nigeria. This study evaluated Girls Invest, an app-based economic empowerment intervention, to address social and economic factors associated with IPV among girls ages 15–19 in low-income communities in Ibadan, Nigeria. Satisfaction, acceptability, and feasibility of the intervention were also assessed. We conducted a 2-armed randomized-controlled trial (16 schools, 8 schools per arm) among girls ages 15–19 (n = 258) to compare the Girls Invest intervention to a wait-list control condition on social and economic risk factors associated with IPV. Quantitative analyses compared survey data at baseline and 6 months follow-up using a difference-in-differences approach. Five focus groups were also conducted among a subset of participants (n = 62) after completing the intervention; focus group data were analyzed and independently coded by two researchers for themes related to participants’ satisfaction, acceptability, and impact on economic and social factors associated with IPV. Participants averaged 16 years old, were predominantly from the Yoruba tribe, and most reported Islam as their religion. Participants reported high levels of acceptability and satisfaction with the app-based intervention. Survey findings indicate statistically significant improvements in gender equitable attitudes (p = 0.02) and non-significant trends toward reduced endorsement of traditional gender roles (p = 0.07) and decreased perception of economic vulnerability (p = 0.20) among Girls Invest participants relative to the waitlist control group. These survey findings were further substantiated by focus group findings with the following themes reported: 1) satisfaction with the app-based platform and financial incentives, 2) greater awareness of issues regarding gender equity and women’s rights, 3) increased awareness of healthy versus unhealthy relationship behaviors, 4) increased knowledge and confidence regarding financial decision-making, and 5) improved expectations for the future regarding vocational opportunities. Notably, participants reported discussing topics they learned with friends and siblings. While a full-scale trial is needed, our findings suggest that Girls Invest holds promise in mitigating social and economic factors associated with IPV. Given the increasing use of smartphones in Nigeria and globally, mobile technology may be a useful platform to support scalability and is lower in cost compared to traditional in-person approaches. Clinical trials unique ID: NCT06942481 (Retrospective registration: 4/9/2025). Data available upon request; study ongoing.

In extreme cold-start scenarios, where users and items have little or no interaction history, traditional recommender systems struggle to generate diverse and meaningful suggestions. This work introduces a unified framework designed to address this challenge using two complementary techniques: (i) a graph augmentation module which densifies the user-item graph using feature-based link prediction, and (ii) a re-ranking algorithm which increases recommendation diversity via centroid-aware sampling across distant user clusters. The resulting architecture also combines graph neural networks for representation learning. We empirically evaluate the framework using a real-world industrial dataset from the telecommunications sector, which is characterized by severe data sparsity and item imbalance. A comprehensive exploration of the possible hyperparameter configurations is conducted, allowing for the assessment of diversity-coverage trade-off and enabling the selection of a configuration that meets specific objectives. The results provide actionable insights for practitioners, demonstrating how to optimize the framework according to specific goals, whether prioritizing diversity, coverage, or computational efficiency. This work serves as a resource for deploying scalable high-quality recommendation systems in real-world industrial settings, addressing the unique challenges posed by extreme cold-start scenarios.

Posttraumatic stress disorder (PTSD) is a severe psychiatric condition associated with significant impairments in daily functioning. Although psychotherapy is the recommended first-line treatment, off-label psychotropic medications are frequently used. This study examined national treatment patterns and disparities in pediatric PTSD care using a large, diverse data set. We conducted a retrospective analysis of pediatric patients (ages: 6 to 18) diagnosed with PTSD (ICD-10: F43.1; N = 61,516) from the TriNetX Research Network. Treatment patterns were analyzed across demographic and clinical variables. Odds ratios (ORs), hazard ratios (HRs), and 95% CIs were calculated. The average age at diagnosis was 11.2 years (SD = 3.67), with females comprising 58.3% of the cohort. Mood disorders (59.2%), ADHD, and anxiety disorders were common comorbidities. Antidepressants were prescribed to 51.4% of patients, most commonly sertraline and fluoxetine. Antipsychotics were used in 30.5% of cases, but only 19.2% had prior psychotherapy. Overall, psychotherapy utilization was 35.5%. Black and Hispanic youth were less likely to receive most psychotropics, and Black youth had lower odds of receiving psychotherapy (OR: 0.85, 95% CI: 0.81-0.89), though antipsychotic use did not differ significantly. After controlling for confounders, greater clinical severity was associated with increased antipsychotic use (aHR: 2.79, 95% CI: 2.68-2.90). This study highlights substantial variability in pediatric PTSD care, including frequent psychotropic use, often preceding psychotherapy, and significant racial and ethnic disparities. These findings emphasize the need for standardized protocols, better access to evidence-based care, and targeted strategies to reduce inequities.

This chapter examines the transformative impact of Artificial Intelligence (AI) on sustainable resource management in the context of mounting global challenges such as population growth, urbanization, and climate change. By emphasizing the importance of innovative AI applications across key sectors such as water resource management, energy optimization, forest and wildlife conservation, and sustainable urban development, while providing clearer transitions to enhance reader comprehension, this chapter showcases the potential benefits of AI in achieving more efficient and equitable resource use. However, it also addresses critical challenges, including data limitations, energy consumption, and ethical considerations that must be navigated. Case studies exemplify successful AI implementations in resource management, integrating recent studies on graphically controlled metric spaces to strengthen the literature review, and discussing the moral and policy frameworks necessary to ensure responsible AI use. As AI continues to evolve, future directions emphasize collaboration and comprehensive integrative strategies to enhance sustainability efforts worldwide.

Smoke detection plays a critical role in preventing fire-related hazards, particularly in intelligent monitoring and early warning systems. Conventional smoke sensors often exhibit limited responsiveness in dynamic environmental conditions, prompting the adoption of IoT-based sensor data combined with machine learning techniques. This study presents a comparative evaluation of four supervised classification algorithms, K-Nearest Neighbors (KNN), Decision Tree, Random Forest, and Gradient Boosting, using the Smoke Detection Dataset from Kaggle. The methodology integrates SMOTE to address class imbalance and Z-score normalization for feature standardization. Hyperparameter tuning was performed using GridSearchCV with 5-fold cross-validation, and model performance was assessed based on accuracy and execution time. Experimental results show that KNN achieved the highest accuracy (98.33%) with the lowest execution time (0.0327 s), whereas Decision Tree recorded the lowest accuracy (84.17%) but remained computationally fast (0.0406 s). Random Forest and Gradient Boosting demonstrated strong predictive capability (97.22% and 96.94%, respectively), but at higher computational costs (1.4338 s and 8.3819 s, respectively). Almost all models achieved perfect scores (1.00) for precision, recall, and F1-score following SMOTE-based balancing, except KNN which obtained slightly lower values (0.99). The findings indicate a trade-off between predictive performance and computational efficiency, suggesting that lightweight models such as KNN are better suited for real-time IoT-based smoke detection. In contrast, ensemble models may be more appropriate for backend analysis. This research contributes an integrated evaluation framework that combines data rebalancing, multi-model benchmarking, and time-based performance analysis, providing practical insights for the development of responsive and scalable early smoke detection systems.

Penelitian ini bertujuan untuk mengetahui bagaimana implementasi teknologi digital dalam proses pembelajaran pendidikan agama Katolik dalam meningkatkan motivasi belajar peserta didik di SMA Swasta Ile boleng di Flores Timur. Implementasi teknologi digital diharapkan mampu meningkatkan motivasi belajar peserta didik. Penelitian ini dilakukan di SMA Swasta Ile Boleng Kabupaten Flores Timur dengan menggunakan metode deskriptif kualitatif. Wawancara dan observasi langsung dilakukan sebagai alat pengumpulan data dan informasi. Penelitian ini dilakukan dari tanggal 07 sampai dengan 14 November 2025. Subyek dalam penelitian ini berjumlah tujuh informan. Hasil penelitian ini mengatakan bahwa teknologi digital dalam pembelajaran PAK di SMA Swasta Ile Boleng sudah berjalan dengan baik, namun belum optimal sehingga hal ini berpengaruh terhadap peningkatan motivasi belajar peserta didik. Walaupun belum optimal karena kekurangan fasilitas pembelajaran, namun peserta didik di sekolah ini memiliki motivasi belajar yang cukup. Sumbangsih dari penelitian ini adalah guru PAK di sekolah ini harus dapat menguasai kompetensi teknologi digital dan fasilitas sekolah yang menunjang kegiatan belajar mengajar perlu diadakan demi meningkatkan motivasi belajar peserta didik.

The current quasi-experimental research tested the effectiveness of synonym-based instruction in stimulating the ESL vocabulary acquisition among the students of secondary level. A total of 60 Grade 9 students were split into an experimental group of students using synonym-centred vocabulary tasks and enhanced using digital tools and a control group of students receiving traditional methods of vocabulary memorization. Paired and independent sample t-tests were used to analyze pretest- posttest vocabulary score, and post-intervention survey was conducted to obtain the perception of students toward the instructional approach. Results showed a statistically significant difference between the performance of the experimental group at the end of the experiment and the control group because of the fact that the instruction based on synonyms allowed achieving a higher level of lexical comprehension, word-wise memory, and contextual application of words. The results of the surveys also indicated a high degree of engagement, motivation, and confidence in students who had been subjected to the strategy, especially in the essay usage of the new vocabulary and in the classroom. In general, the research proves the overall positive cognitive and affective benefits of synonym-based instruction in comparison to traditional vocabulary instruction and has a high potential of supplementing ESL instruction at the high school level.

In response to the efficiency bottleneck in the overall design phase of artillery, this paper proposes an agile design tool for artillery based on the RecurDyn solver, IAOD. This tool achieves rapid evaluation and optimization design of artillery dynamics simulation models through parametric modeling and integration of shooting test data. The IAOD system adopts a fully parameterized driven template model, combined with machine learning, parameter identification, and multi-level optimization algorithms, significantly improving the design efficiency of large caliber self-propelled artillery. The system innovatively applied a sparse single hidden layer neural network proxy model and a simulation parameter identification method based on test data, achieving multi-objective collaborative optimization. The effectiveness and practicality of the IAOD system have been verified through the practical application of a vehicle mounted artillery design case, demonstrating its potential for application in the field of artillery design. This study has significant military and defense value in improving the efficiency and quality of artillery equipment development.