A5004798067- EEG and Brain-Computer Interfaces
- Topic Modeling
- Recommender Systems and Techniques
- Privacy-Preserving Technologies in Data
- Natural Language Processing Techniques
- Second Language Acquisition and Learning
- Sleep and Work-Related Fatigue
- Human Mobility and Location-Based Analysis
- EFL/ESL Teaching and Learning
- Social Robot Interaction and HRI
- Chronic Disease Management Strategies
- Blind Source Separation Techniques
- Sentiment Analysis and Opinion Mining
- Machine Learning in Healthcare
- Myxozoan Parasites in Aquatic Species
- Emotion and Mood Recognition
- Marine and coastal ecosystems
- Aquatic life and conservation
- Sensor Technology and Measurement Systems
- Coastal and Marine Dynamics
- Statistical Methods and Inference
- Microbial infections and disease research
- Advanced Graph Neural Networks
- Crustacean biology and ecology
- Fire Detection and Safety Systems
Ningbo University
2022-2023
Hong Kong Baptist University
2023
East China University of Science and Technology
2014-2016
Yunnan Normal University
2006-2011
Objective: Daytime short nap involves individual physiological states including alertness and drowsiness. In order to have a better understanding of the periodical rhymes then promote good interpretability alertness, aim this study is detect drowsiness during daytime nap. Methods: A method Bayesian-copula discriminant classifier (BCDC) was introduced based on features extracted from electroencephalogram (EEG) signals. As an extension traditional Bayesian decision theory, BCDC tries construct...
Large Language Models (LLMs) like ChatGPT and GPT-4 are versatile capable of addressing open-domain question-answering(QA) tasks effectively. However, general LLMs, which developed on data, may lack the domain-specific knowledge essential for in vertical domains, such as legal, medical, etc. To address this issue, previous approaches either conduct continuous pre-training with data or employ retrieval augmentation to support LLMs handling QA tasks. Unfortunately, these strategies...
Daytime short nap involves physiological processes, such as alertness, drowsiness and sleep. The study of the relationship between based on signals is a great way to have better understanding periodical rhymes states. A model Bayesian nonnegative CP decomposition (BNCPD) was proposed extract common multiway features from group-level electroencephalogram (EEG) signals. As an extension decomposition, BNCPD prior distributions factor matrices, while underlying rank could be determined...
Large Language Models (LLMs) like ChatGPT and GPT-4 are versatile capable of addressing a diverse range tasks. However, general LLMs, which developed on open-domain data, may lack the domain-specific knowledge essential for tasks in vertical domains, such as legal, medical, etc. To address this issue, previous approaches either conduct continuous pre-training with data or employ retrieval augmentation to support LLMs. Unfortunately, these strategies cost-intensive unreliable practical...
Kudoa (Myxosporea: Multivalvulida) parasites are critical pathogens in marine and freshwater fish associated with significant economic losses reduced market prices caused by post-mortem myoliquefaction or numerous cysts on muscles. In the present study, large yellow croakers infected were found during disease surveillance China November 2020 used for morphological observation characterization using light DIC microscopy electron microscopy. Numerous creamy-white oval plasmodia observed...
Point of Interest (POI) recommendation system rec-ommends places in which users have never been to but may be in-terested. Traditionally, it centrally collects contextual information and interaction data model users' preferences, raises many privacy concerns. The current studies habitually sacrifice the performance cope with con-cerns. To protect while ensuring POI system, we propose a Ranking-based Federated Recommendation Geographic Effect (RFPG). RFPG allows reserve their private on local...
Predictive analysis of electronic health records (EHR) is a fundamental task that could provide actionable insights to help clinicians improve the efficiency and quality care. EHR are commonly recorded in binary format contain inevitable missing data. The nature missingness may vary by patients, clinical features, time, which incurs observation bias. It essential account for bias or predictive performance be substantially compromised. In this paper, we develop propensity-adjusted temporal...
Abstract Enterocytozoon hepatopenaei (EHP) is an important factor limiting the growth of Penaeus vannamei. Its simple transmission route and lack effective control drugs often lead to great losses for shrimp aquaculture industry. In this study, some basic information EHP was provided, including spore purification, germination promoting treatment, quantification determination key genes expression, activities enzymes in glycolysis pathway at stage germination. The result showed that light...
The cross-domain sequential recommendation aims to predict the next item based on a sequence of recorded user behaviours in multiple domains. We propose novel Cross-domain Sequential Recommendation approach with Graph-Collaborative Filtering (CsrGCF) alleviate sparsity issue user-interaction data. Specifically, we design time-aware and relation-aware graph attention mechanisms collaborative filtering exploit high-order behaviour patterns users for promising results both Time-aware Graph...
The sEMG (Surface Electromyogram) is a kind of physiological signal corresponding to the muscle action. analysis can obtain charateristics motion information. In this paper, remote control system based on was developed. consists host and client computer. A recording device designed. recorded data transit by Zigbee wireless communication technique obtained analysed developed multivariable ANFIS model in order recognize hand movement. recognition result shown computer local area network....