Wuhan Business University

This paper suggests a Bidirectional LSTM (ELCNN-BiLSTM) model, which is an E-Commerce Logistics CNN-based model of intelligent decision-making and timeliness prediction in the cross-border logistics processes. The data to be used in the study was gathered within a global e-commerce logistics system of a cosmetics retailer with 1,25, 000 delivery records that have 24 spatio-temporal and operational characteristics, such as the shipment weight, their origin, destination, the distance of the route, the time in customs, and the time in the warehouse. The formulated task is a binary classification task to forecast whether a delivery is on-time or delayed, with an on-time threshold parameter being delivery within +/-1 day of the expected arrival. Preprocessing was performed on the basis of alignment of the timestamps, normalisation, treatment of 3.4% missing values by interpolation, and resampling to address a 1:1.7 ratio of the class imbalance with SMOTE. The ELCNN-BiLSTM architecture proposed is a combination of spatial feature extraction with convolutional layers and long-term temporal dependencies with the bidirectional LSTM layers that allow adaptive acquisition of logistics patterns on routes and time zones. Benchmarking of the model was done against LSTM, CNN-LSTM, Att-GRU, and XGBoost baselines on measurements of Accuracy, Precision, Recall, F1-Score, PR-AUC and ROC-AUC. Findings reveal that the suggested model has a better predictive power with 95.4 % accuracy, 93.9 % precision and 96.3 % of ROC-AUC, which are significantly higher than other approaches in all the most important measures. The ability and robustness of the model to generalise and statistical significance tests were statistically proven through statistical tests and k-fold cross-validation. The study presents a powerful data-oriented model of predicting supply chain delays in real-time, which can improve the reliability of supply chains and help to make intelligent decisions in the global e-commerce business.

Introduction The increasing complexity and non-linearity of financial markets make traditional linear models inadequate for systemic risk assessments. This study aims to develop a new framework for identifying outlier events and critical transitions in financial markets. Methods We propose a framework that combines a complex network approach with visibility graph algorithms. First, a comprehensive financial stress index (FSI) for China is constructed by integrating sub-indices from the bond, stock, money, and foreign exchange markets, along with time-varying cross-market correlations. The FSI and sub-index time series are converted into complex networks via the visibility graph algorithm. A Rayleigh-entropy-based overlapping influence algorithm is introduced to detect critical risk nodes, addressing node interdependencies and network loops overlooked by traditional percolation theories. The framework is validated using daily data from January 2015 to June 2025, with cross-market stress transmission examined through a vector autoregression model. Results The approach effectively identifies major financial stress periods and cross-market stress transmission paths. A stock market shock raises the bond market stress index by 0.08 within five trading days. A reserve requirement ratio cut by the People’s Bank of China reduces the average money market stress index by 0.12 within 30 days. Stress originates and spreads across sub-markets with measurable time lags, and the bond market credit spread is the dominant driver of stochastic fluctuations in the FSI. Discussion The proposed network-based method, paired with model-driven transmission analysis, serves as a robust dynamic tool for monitoring and early warning of financial systemic risk. It provides data-supported insights for both academic researchers and policymakers.

In recent years, deep learning-based multimodal emotion recognition methods have been widely applied in student mental health monitoring. However, main-stream models still exhibit limitations in feature coupling and temporal sequence modeling. To address these challenges, this study proposes an enhanced residual spatiotemporal attention network that integrates multi-dimensional perceptual features with residual connection mechanisms, focusing on improving the spatial recognition of key emotional cues and preserving temporal dependencies. An intervention experiment was conducted among a real-world cohort of university students. The results show that the proposed model achieved recognition accuracy above 83% for depression, anxiety, and stress categories. Participants in the intervention group also demonstrated significant improvements in health aware-ness and behavioral performance, with Cohen's d effect sizes generally exceeding 1.2. The findings suggest that the model is well-suited for dynamic emotion recognition and cognitive guidance tasks in university mental health intervention scenarios. This study contributes a viable technical approach and model frame-work for enabling personalized and adaptive mental health interventions in higher education settings.

Heterojunction photocatalysis holds great significance for low-cost and efficient environmental remediation processes, particularly for addressing persistent antibiotic contamination. Here, BiO 2-X -Bi 2 O 2 CO 3 heterojunction photocatalysts are fabricated via a low-temperature solvothermal epitaxial growth method. The intimate interfacial contact between the BiO 2-X nanoparticles and the epitaxially grown Bi 2 O 2 CO 3 nanosheets was detected, which endows the heterostructure with significantly enhanced visible-light activity. The photocatalytic properties of the optimized composite, BiO 2-X -Bi 2 O 2 CO 3 -20, are investigated in detail toward tetracycline (TC) degradation. The superior charge carrier dynamics is confirmed to be vital to enhanced performance, as evidenced by PL spectroscopy and transient photocurrent analysis, which collectively indicate that the heterojunction effectively suppresses electron-hole recombination and promotes charge transfer. Mechanistic studies, including ESR and radical trapping experiments, validated a Z-scheme charge transfer pathway, which ensures the preservation of the strong reducing (e - at -0.75 eV) and oxidizing (h + at 2.51 eV) potentials. This preserved high energy (∼3.26 eV) promotes the generation of the superoxide radical (·O 2 - ), conclusively identified as the dominant active species responsible for the high efficiency. This work introduces a robust Z-scheme method for Bi-based photocatalysts, which is ready to extend to other heterogeneous systems and offers a new option to design high-performance catalysts for efficient antibiotic-contaminated wastewater treatment under visible light. The BiO 2-X -Bi 2 O 2 CO 3 heterojunction with built-in internal electric field (IEF) efficiently separates photogenerated charge carriers under visible light, generating ·O 2 - and ·OH radicals for significantly enhanced photocatalytic degradation of organic pollutants. • Epitaxial BiO 2-x -Bi 2 O 2 CO 3 synthesis boosts interface charge transfer. • Z-scheme validated by Mott-Schottky and ESR spectroscopy. • 15 times faster tetracycline degradation than pure Bi 2 O 2 CO 3 component. • Preserved 3.26 eV potential maximizes ·O 2 - generation. • Robust Z-scheme resolves Bi-catalyst light/power trade-off constraint.