A5082275076- Recommender Systems and Techniques
- Neurological disorders and treatments
- Complex Network Analysis Techniques
- Human Mobility and Location-Based Analysis
- Advanced Graph Neural Networks
- Parkinson's Disease Mechanisms and Treatments
- Surgical Simulation and Training
- Functional Brain Connectivity Studies
- Data-Driven Disease Surveillance
- Expert finding and Q&A systems
- Psychosomatic Disorders and Their Treatments
- Voice and Speech Disorders
- Generative Adversarial Networks and Image Synthesis
- Artificial Intelligence in Healthcare and Education
- Impact of Light on Environment and Health
- Neural dynamics and brain function
- Migraine and Headache Studies
- Acupuncture Treatment Research Studies
- Stochastic Gradient Optimization Techniques
- Pharmacological Effects of Natural Compounds
- EEG and Brain-Computer Interfaces
- Sparse and Compressive Sensing Techniques
- Video Coding and Compression Technologies
- Advanced Bandit Algorithms Research
- Domain Adaptation and Few-Shot Learning
Durham University
2022-2025
King's College London
2023
University of Southampton
2020-2021
Dongzhimen Hospital Affiliated to Beijing University of Chinese Medicine
2019-2020
Beijing University of Chinese Medicine
2019-2020
Huawei Technologies (China)
2013
IBM (United States)
2010-2011
IBM Research (China)
2008-2010
Accurately capturing user preferences over time is a great practical challenge in recommender systems. Simple correlation typically not meaningful, since users change their due to different external events. User behavior can often be determined by individual's long-term and short-term preferences. How represent users' preferences? leverage them for temporal recommendation? To address these challenges, we propose Session-based Temporal Graph (STG) which simultaneously models time. Based on...
Considering the natural tendency of people to follow direct or indirect cues other people's activities, collaborative filtering-based recommender systems often predict utility an item for a particular user according previous ratings by similar users. Consequently, effective searching most related neighbors is critical success recommendations. In recent years, tagging with social bookmarking as their key component from suite Web 2.0 technologies allow users freely bookmark and assign semantic...
Total variation (TV) regularization has important applications in signal processing including image denoising, deblurring, and reconstruction. A significant challenge the practical use of TV lies nondifferentiable convex optimization, which is difficult to solve especially for large-scale problems. In this paper, we propose an efficient alternating augmented Lagrangian method (ADMM) total The proposed algorithm applicable tensors, thus it can multidimensional One appealing feature that does...
Object . The purpose of this study was to fully assess the role statins in primary prevention coronary heart disease (CHD). Methods We searched six databases (PubMed, Cochrane Library, Web Science, China National Knowledge Infrastructure, Wanfang Database, and Chinese Scientific Journal Database) identify relevant randomized controlled trials (RCTs) from inception 31 October 2017. Two review authors independently assessed methodological quality analysed data using Rev Man 5.3 software. Risk...
Multi-label classification, or the same example can belong to more than one class label, happens in many applications. To name a few, image and video annotation, functional genomics, social network annotation text categorization are some typical Existing methods have limited performance both efficiency accuracy. In this paper, we propose an extension over decision tree ensembles that handle challenges. We formally analyze learning risk of Random Decision Tree (RDT) derive upper bound is...
Visualizing massive scale human movement in cities plays an important role solving many of the problems that modern face (e.g., traffic optimization, business site configuration). In this article, we study a big mobile location dataset covers millions city residents, but is temporally sparse on trajectory individual user. Mapping trajectories to illustrate population poses several challenges from both analysis and visualization perspectives. literature, there are few techniques designed for...
Content-centric social websites, such as discussion forums and blog sites, have flourished during the past several years. These sites often contain overwhelming amounts of information that are also being updated rapidly. To help users locate their interests at (e.g., interesting blogs to read or join), researchers developed a number recommendation technologies. However, it is difficult make effective recommendations for new (a.k.a. cold start problem) due lack user preferences interests)....
Content-centric social Web sites, such as discussion forums and blog have flourished during the past several years. These sites often contain overwhelming amounts of information that are also being updated rapidly. To help users locate their interests at (e.g., interesting blogs to read or join), researchers developed a number recommendation technologies. However, it is difficult make effective recommendations for new (a.k.a. cold start problem) due lack user preferences interests)....
Abstract Purpose Current methods for diagnosis of PD rely on clinical examination. The accuracy ranges between 73 and 84%, is influenced by the experience assessor. Hence, an automatic, effective interpretable supporting system symptom identification would support clinicians in making more robust diagnostic decisions. Methods We propose to analyze Parkinson’s tremor (PT) analysis PD, since PT one most typical symptoms with broad generalizability. To realize idea, we present SPA-PTA, a deep...
Abstract Health professional education stands to gain substantially from collective efforts toward building video databases of skill performances in both real and simulated settings. An accessible resource videos that demonstrate an array – good bad—provides opportunity for interdisciplinary research collaborations can advance our understanding movement reflects technical expertise, support educational tool development, facilitate assessment practices. In this paper we raise important...
Depersonalization-Derealization Disorder (DPD), a prevalent psychiatric disorder, fundamentally disrupts self-consciousness and could significantly impact the quality of life those affected. While existing research has provided foundational insights for this limited exploration brain dynamics in DPD hinders deeper understanding its mechanisms. It restricts advancement diagnosis treatment strategies. To address this, our study aimed to explore DPD.
User-based collaborative filtering methods typically predict a user's item ratings as weighted average of the given by similar users, where weight is proportional to user similarity. Therefore, accuracy similarity key success recommendation, both for selecting neighborhoods and computing predictions. However, computed similarities between users are somewhat inaccurate due data sparsity.
Cities are living systems where urban infrastructures and their functions defined evolved due to population behaviors. Profiling the cities functional regions has been an important topic in design planning. This paper studies a unique big data set which includes daily movement of tens millions city residents, develop visual analytics system, namely UrbanFACET, discover visualize dynamical profiles multiple residents. user set, acquired from mobile users' agnostic check-ins at thousands phone...
The treatment of pain with complementary and alternative medicine has gradually attracted international attention. Traditional Chinese medicine(TCM) clear composition mechanism will become a new starting point for drug research development. Siegesbeckiae Herba is commonly used TCM herb in the rheumatic arthralgia. With chemical composition, long history clinical application certain modern basis chronic pain. It good development prospects analgesia. Based on occurrence principle known action...
Surgical workflow anticipation can give predictions on what steps to conduct or instruments use next, which is an essential part of the computer-assisted intervention system for surgery, e.g. reasoning in robotic surgery. However, current approaches are limited their insufficient expressive power relationships between instruments. Hence, we propose a graph representation learning framework comprehensively represent instrument motions surgical problem. In our proposed representation, maps...