A5103209723- Topic Modeling
- Natural Language Processing Techniques
- Biomedical Text Mining and Ontologies
- Advanced Text Analysis Techniques
- Machine Learning in Healthcare
- Stock Market Forecasting Methods
- Sentiment Analysis and Opinion Mining
- Advanced battery technologies research
- Complex Systems and Time Series Analysis
- Advanced Battery Materials and Technologies
- Advancements in Battery Materials
- Text and Document Classification Technologies
- Handwritten Text Recognition Techniques
- Text Readability and Simplification
- Artificial Intelligence in Healthcare
- Computational and Text Analysis Methods
- Time Series Analysis and Forecasting
- Neural Networks and Applications
- Semantic Web and Ontologies
- Boron and Carbon Nanomaterials Research
- Medical Coding and Health Information
- Fuzzy Logic and Control Systems
- Rough Sets and Fuzzy Logic
- Recommender Systems and Techniques
- GaN-based semiconductor devices and materials
Central South University
2021-2024
Wuhan University of Science and Technology
2023-2024
Henan University
2023-2024
Hunan Communications Research Institute
2021
Harbin Institute of Technology
2014-2019
Chinese Academy of Medical Sciences & Peking Union Medical College
2015
University of North Dakota
2002
Peking University
1997
It has been shown that news events influence the trends of stock price movements.However, previous work on news-driven market prediction rely shallow features (such as bags-of-words, named entities and noun phrases), which do not capture structured entity-relation information, hence cannot represent complete exact events.Recent advances in Open Information Extraction (Open IE) techniques enable extraction from web-scale data.We propose to adapt IE technology for event-based movement...
Medical knowledge graphs (MKGs) are the basis for intelligent health care, and they have been in use a variety of medical applications. Thus, understanding research application development MKGs will be crucial future relevant biomedical field. To this end, we offer an in-depth review MKG work. Our begins with examination four types information sources, graph creation methodologies, six major themes development. Furthermore, three popular models reasoning from viewpoint discussed. A...
Metallic zinc electrode with a high theoretical capacity of 820 mAh g-1 is highly considered as promising candidate for next-generation rechargeable batteries. However, the unavoidable hydrogen evolution, uncontrolled dendrite growth, and severe passivation reaction badly hinder its practical implementations. Herein, robust polymer-alloy artificial protective layer designed to realize dendrite-free Zn metal anode by integration zincophilic SnSb nanoparticles Nafion. In comparison bare...
Abstract A 3D nanostructured scaffold as the host for zinc enables effective inhibition of anodic dendrite growth. However, increased electrode/electrolyte interface area provided by using matrices exacerbates passivation and localized corrosion Zn anode, ultimately bringing about degradation electrochemical performance. Herein, a nanoscale coating inorganic–organic hybrid (α‐In 2 Se 3 ‐Nafion) onto flexible carbon nanotubes (CNTs) framework (ISNF@CNTs) is designed plating/stripping to...
Social media platforms are often used by people to express their needs and desires. Such data offer great opportunities identify users’ consumption intention from user-generated contents, so that better tailored products or services can be recommended. However, there have been few efforts on mining commercial intents social contents. In this paper, we investigate the use of intentions for individuals. We develop a Consumption Intention Mining Model (CIMM) based convolutional neural network...
Xiao Ding, Kuo Liao, Ting Liu, Zhongyang Li, Junwen Duan. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.
There is mounting evidence that the subcellular localization of lncRNAs can provide valuable insights into their biological functions. In real world transcriptomes, are usually localized in multiple localizations. Furthermore, have specific patterns for different Although several computational methods been developed to predict lncRNAs, few them designed localizations, and none take motif specificity consideration.In this study, we proposed a novel deep learning model, called LncLocFormer,...
Determining drug-drug interactions (DDIs) is an important part of pharmacovigilance and has a vital impact on public health. Compared with drug trials, obtaining DDI information from scientific articles faster lower cost but still highly credible approach. However, current text extraction methods consider the instances generated to be independent ignore potential connections between different in same article or sentence. Effective use external data could improve prediction accuracy, existing...
Chinese electronic medical records (EMR) presents significant challenges for named entity recognition (NER) due to their specialized nature, unique language features, and diverse expressions. Traditionally, NER is treated as a sequence labeling task, where each token assigned label. Recent research has reframed within the machine reading comprehension (MRC) framework, extracting entities in question-answer format, achieving state-of-the-art performance. However, these MRC-based methods have...
Retrieval-Augmented Large Language Models (LLMs), which integrate external knowledge into LLMs, have shown remarkable performance in various medical domains, including clinical diagnosis. However, existing RAG methods struggle to effectively assess task difficulty make retrieval decisions, thereby failing meet the requirements for balancing efficiency and accuracy. So this paper, we propose FIND (\textbf{F}ine-grained \textbf{In}formation \textbf{D}ensity Guided Adaptive RAG), a novel...
Electronic Medical Records (EMRs), while integral to modern healthcare, present challenges for clinical reasoning and diagnosis due their complexity information redundancy. To address this, we proposed medIKAL (Integrating Knowledge Graphs as Assistants of LLMs), a framework that combines Large Language Models (LLMs) with knowledge graphs (KGs) enhance diagnostic capabilities. assigns weighted importance entities in medical records based on type, enabling precise localization candidate...
Alzheimer's Disease (AD) is a neurodegenerative disorder that significantly impacts patient's ability to communicate and organize language. Traditional methods for detecting AD, such as physical screening or neurological testing, can be challenging time-consuming. Recent research has explored the use of deep learning techniques distinguish AD patients from non-AD by analysing spontaneous speech. These models, however, are limited availability data. To address this, we propose novel...
Event Extraction (EE) is a key task in information extraction, which requires high-quality annotated data that are often costly to obtain. Traditional classification-based methods suffer from low-resource scenarios due the lack of label semantics and fine-grained annotations. While recent approaches have endeavored address EE through more data-efficient generative process, they overlook event keywords, vital for EE. To tackle these challenges, we introduce KeyEE, multi-prompt learning...
The spread of misinformation on social media is a serious issue that can have negative consequences for public health and political stability. While detecting identifying be challenging, many attempts been made to address this problem. However, traditional models focus pairwise relationships propagation paths may not effective in capturing the underlying connections among multiple tweets. To limitation, proposed “Conversation-Branch-Tweet” hypergraph convolutional network (CBT-HGCN) uses...
In general, physicians make a preliminary diagnosis based on patients' admission narratives and conditions, largely depending their experiences professional knowledge. An automatic accurate tentative clinical would be of great importance to physicians, particularly in the shortage medical resources. Despite its value, little work has been conducted this method. Thus, study, we propose fusion model that integrates semantic symptom features contained text. The input text are initially captured...