Applied Research, Volume 5, Issue 5, October 2026.This study integrates meso‐void characterization, statistical analysis, and machine learning to predict the sound absorption of porous asphalt concrete. Void‐ratio‐related features dominate prediction, while Extra Trees performs best among five evaluated models, supporting acoustic‐oriented PAC mixture design.
Yizhe Han +6 more
wiley +1 more source
A short-term postoperative prediction model for colorectal cancer using comprehensive geriatric assessment. [PDF]
BMC GeriatrYang D +5 more
europepmc +1 more source
Low-cost IoT real-time environmental monitoring and one-hour LSTM forecasting for small-scale agricultural applications. [PDF]
Sci RepHebishy GH +3 more
europepmc +1 more source
Artificial Intelligence‐Driven Traffic Flow Forecasting and Spatial Road Operation Risk Assessment
Applied Research, Volume 5, Issue 5, October 2026.The AGWN TF framework integrates adaptive graph‐based spatiotemporal forecasting with explicit temporal feature fusion for multistep traffic flow prediction and spatial road operation assessment. The framework improves forecasting performance, particularly at middle and longer prediction horizons, while the proposed Road Operation Risk Index converts ...
Yifei Zhu, Tongyao Jin, Hanzheng Sun
wiley +1 more source
Empirical Determination and Modelling of Compressive Strength of Cement Composites with Waste Tyre Rubber. [PDF]
Materials (Basel)Samardzioska T +4 more
europepmc +1 more source
Applied Research, Volume 5, Issue 5, October 2026.A low‐cost Python‐based machine vision framework quantified rice morphology and optimized analog‐rice extrusion. Within the tested range, 50 Hz screw speed and 45 Hz cutter speed produced grains with 96.6% morphological similarity to the commercial Pusa Basmati 1121 reference.
Rosniza Rabilah +2 more
wiley +1 more source
Validation of heart rate measurements from the polar verity sense during the submaximal Ekblom-Bak test in adult individuals with obesity. [PDF]
J Rehabil MedOnstad GR, Lohne FK, Fimland MS.
europepmc +1 more source
Applied Research, Volume 5, Issue 5, October 2026.This critical review synthesizes ML‐enabled occupancy behaviour modelling across sensing approaches, prediction methods and building contexts. Emerging advances in non‐intrusive sensing, hybrid and privacy‐preserving learning show strong potential, while transferability, heterogeneous data and inconsistent benchmarking remain key barriers to scalable ...
Norhafiza Mohamad +4 more
wiley +1 more source
Data-Driven Inverse Design of Carbon Fibre-Reinforced Polymer Laminated Plates via a Tandem Neural Network Framework. [PDF]
Polymers (Basel)Huang M, Yuan L, Ran J, Liu H, Huang Y.
europepmc +1 more source
A Mechanism for Analyzing the Energy Consumption and Environmental Impact of LoRa Sensors Over Time
International Journal of Communication Systems, Volume 39, Issue 15, October 2026.An SPN‐based mechanism, validated against NS‐3 with MAPE below 0.3%, reveals how packet generation rate and spreading factor shape LoRa sensor lifetime, energy consumption, and operational CO2$$ {}_2 $$ emissions, supporting sustainable planning of large‐scale IoT deployments. ABSTRACT Long Range (LoRa) sensors are widely adopted in Internet of Things (
Vandirleya Barbosa +5 more
wiley +1 more source