A5101598635- Stock Market Forecasting Methods
- Financial Markets and Investment Strategies
- Recommender Systems and Techniques
- Atmospheric chemistry and aerosols
- Advanced Bandit Algorithms Research
- Explainable Artificial Intelligence (XAI)
- Phonocardiography and Auscultation Techniques
- Context-Aware Activity Recognition Systems
- Human Mobility and Location-Based Analysis
- Online Learning and Analytics
- Computational Physics and Python Applications
- Market Dynamics and Volatility
- Per- and polyfluoroalkyl substances research
- Image Retrieval and Classification Techniques
- Vehicle emissions and performance
- Advanced Computational Techniques and Applications
- Toxic Organic Pollutants Impact
- Air Quality and Health Impacts
- Engineering and Test Systems
- Intelligent Tutoring Systems and Adaptive Learning
- ECG Monitoring and Analysis
- Fault Detection and Control Systems
China University of Geosciences
2025
Huazhong University of Science and Technology
2017-2022
University of Science and Technology
2022
Southeast University
2012
Sun Yat-sen University
2010
An intensive field experiment was conducted at an urban and a rural site in Hong Kong to identify the influence of biomass burning emissions transported from distinct regions on ambient aerosol coastal southeast China. Watersoluble ionic carbonaceous species, specifically tracer levoglucosan, were analysed. Elevated levoglucosan concentrations with maxima 91.5 133.7 ng m-3 overall average 30 36 observed sites, respectively. By combining analysed meteorological data, backward trajectories,...
Abstract In this study, a detailed analysis of 20 per- and polyfluoroalkyl substances (PFAS) was conducted, in different environmental media the Fu River, main sewage storage body, located near Tianhe Airport Wuhan, China. The PFAS included 13 perfluorocarboxylic acids (C4–C18), four perfluorosulfonic (C4, C6, C8, C10), three PFAS. surface water samples, short-chain perfluorobutanesulfonic acid (PFBS) perfluorobutanoic were most prevalent highest concentrations, 168 ng/L 49.7 ng/L,...
Financial portfolio management is the process of periodically reallocating a fund into different financial investment products, with goal achieving maximum profits. While conventional machine learning methods try to predict price trends, reinforcement based makes trading decisions according changes directly. However, existing are limited in extracting change information at single-scale level, which their performance still not satisfactory. In this paper, inspired by Inception network that...
Reinforcement learning-based portfolio management has recently attracted extensive attention. However, deep reinforcement learning methods are unexplainable and considered to be potentially risky, difficult trusted regulated by users. To address these problems, we propose an eXplainable framework for Portfolio Management, named XPM, which is efficient, concise, can provide faithful explanations network outputs. Specifically, first design a policy management, uses temporal convolutional (TCN)...
The explosive growth of e-commerce and online service has led to the development recommender system. Aiming provide a list items meet user’s personalized need by analyzing his/her interaction 1 history, system been widely studied in academic industrial communities. Different from conventional systems, sequential systems attempt capture pattern users’ behaviors evolution preferences. Most existing recommendation models only focus on user sequence, but neglect item sequence. An sequence also...