---

Title: Multiscale Transfer Learning for Low-Resource Named Entity Recognition in Administrative Texts in Emerging Economies

Abstract:Local-language administrative corpora lack labeled entities. Cross-lingual transfer with domain adapters improves place and organization recognition relative to scratch training.

---

---

Title: Integrated Microclimate Mapping of Urban Pocket Parks for Heat-Island Mitigation Across Heterogeneous Landscapes

Abstract:Small parks are promoted for amenity value, yet cooling contribution is rarely quantified. Dense logger grids show evening temperature reductions scale with continuous leaf area more than ornamental design alone.

---

---

Title: Empirical Graph Neural Networks for Predicting Cascading Failures in Regional Power Grids for Decision Support

Abstract:N-1 contingency lists miss multi-hop failure paths. Graph neural models trained on simulated outages improve early ranking of critical lines relative to classical loading metrics.

---

---

Title: Comparative Gamified Formative Assessment Effects on Motivation in Secondary Mathematics Under Resource Constraints

Abstract:Digital badges sustain practice, yet evidence on conceptual learning remains mixed. Comparing gamified quizzes with worksheets, motivation and algebra scores improve when badges are tied to mastery rather than speed alone.

---

---

Title: Soil Moisture Sensor Networks for Deficit Irrigation Scheduling in Young Plantings

Abstract:Young plantings risk over-irrigation when schedules ignore root-zone heterogeneity. Multi-depth capacitance probes guiding deficit regimes maintain shoot growth while cutting applied water in the driest months.

---

---

Title: Regional Stochastic Programming for Vaccine Outreach Routes Under Cold-Chain Failures With Uncertainty Quantification

Abstract:Rural immunization campaigns lose doses when coolers fail. Stochastic routing with failure scenarios reallocates spare capacity and reduces expected spoiled vials versus static tours.

---

---

Title: Experimental Causal Inference of Congestion Pricing Effects on Retail Footfall Using Synthetic Controls Using Open Data Streams

Abstract:Retailers fear pricing zones will empty streets. Synthetic control estimates of footfall around zone boundaries separate pricing effects from concurrent retail cycles.

---

---

Title: Longitudinal Edge Computing Pipelines for Real-Time Greenhouse Microclimate Anomaly Detection During Seasonal Transitions

Abstract:Cloud-only analytics introduce latency that delays response to ventilation and irrigation faults. We deploy lightweight anomaly detectors on edge gateways fed by temperature, humidity, and CO2 streams. Edge pipelines cut mean detection delay while preserving interpretable alerts under intermittent connectivity.

---

---

Title: Explainable Gradient Boosting for Early Warning of Hospital-Acquired Pressure Injury Using Nursing Assessment Scores

Abstract:Pressure injury prevention bundles rely on periodic Braden scores that may miss rapid deterioration in mobilized postoperative patients. We train gradient-boosted models on nursing documentation, mobility logs, and laboratory trends and pair predictions with SHAP attributions for bedside nurses. The model flags high-risk patients one shift earlier than score-only rules in external validation, and ward staff report that explanations align with observed mobility and perfusion changes.

---

---

Title: Federated Learning for Multicenter Prediction of Acute Kidney Injury After Cardiac Surgery Without Centralized Patient Data

Abstract:Hospitals hesitate to pool perioperative records for AKI models because of privacy rules and heterogeneous EHR schemas. We train a federated gradient-boosted ensemble across seven cardiac centers that never exchange raw features, synchronizing only encrypted parameter updates. External validation on a held-out site yields AUROC comparable to a centrally trained baseline while reducing false-positive alerts that drive unnecessary nephrology consults during the first 48 postoperative hours.

---