A5027800550- Topic Modeling
- Biomedical Text Mining and Ontologies
- Advanced Text Analysis Techniques
- Natural Language Processing Techniques
- Gene expression and cancer classification
- Bioinformatics and Genomic Networks
- Anomaly Detection Techniques and Applications
- Machine Learning in Bioinformatics
- Video Surveillance and Tracking Methods
- Metabolomics and Mass Spectrometry Studies
- Text and Document Classification Technologies
- Privacy-Preserving Technologies in Data
- Gait Recognition and Analysis
- Data Quality and Management
- Image and Object Detection Techniques
- AI in cancer detection
- Complex Network Analysis Techniques
- Single-cell and spatial transcriptomics
- Gene Regulatory Network Analysis
- Recommender Systems and Techniques
- Advanced Graph Neural Networks
- Cellular Automata and Applications
- Robotic Path Planning Algorithms
- Cell Image Analysis Techniques
- Rough Sets and Fuzzy Logic
Liaoning Normal University
2019-2025
Dalian University of Technology
2016-2024
Dalian University
2023-2024
Shanghai University
2023-2024
Xiamen University of Technology
2023
Wuhan Textile University
2023
Xiamen University
2021-2023
Xidian University
2021
Abstract Drowning is a significant public health concern. A video drowning detection algorithm helpful tool for finding victims. However, there are three challenges that research typically encounters: lack of actual data, subtle early traits, and real time. In this paper, the authors propose an underwater computer vision based device composed embedded AI devices, camera, waterproof case to solve above problems. The utilizes high‐performance computing Jetson Nano realize real‐time events...
Extracting biomedical events from literature plays an important role in the field of text mining, and trigger detection is a key step event extraction. We propose two-stage method for detection, which divides into recognition stage classification stage, different features are selected each stage. In first we select more suitable recognition, second that helpful to adopted. Furthermore, integrate word embeddings represent words semantically syntactically. On multi-level extraction (MLEE)...
Abstract Background Biomedical event extraction is a fundamental task in biomedical text mining, which provides inspiration for medicine research and disease prevention. events include simple complex events. Existing methods usually deal with uniformly, the performance of relatively low. Results In this paper, we propose fine-grained Bidirectional Long Short Term Memory method extraction, designs different argument detection models respectively. addition, multi-level attention designed to...
Each bank has different clients and each client may have transactions with multiple banks. Hence, clients' data in a single be partial incomplete. If the can combined, obtains comprehensive information, so as to better carry out business enhance quality of service, such recommending financial products inquiring about personal credit records. However, after promulgation GDPR by European Union 2018, it is illegal directly consolidate crossing enterprises due privacy security concerns,...
Cellular networks realize their functions by integrating intricate information embedded within local structures such as regulatory paths and feedback loops. However, the precise mechanisms of how topologies determine global network dynamics induce bifurcations remain unidentified. A critical step in unraveling integration is to identify governing principles, which underlie flow. Here, we develop cumulative linearized approximation (CLA) algorithm address this issue. Based on perturbation...
As the crucial and prerequisite step in biomedical event extraction, trigger detection has attracted much attention. Most of existing methods either rely on elaborately designed features or consider only within a window. Another challenge is that treat each word sentence equally. Also, most ignore sentence-level semantic information. Therefore, we propose method based Bidirectional Long Short Term Memory (BiLSTM) neural network, which can skip manual complex feature extraction. Furthermore,...
Biomedical event extraction is an important and challenging task in Information Extraction, which plays a key role for medicine research disease prevention. Most of the existing detection methods are based on shallow machine learning mainly rely domain knowledge elaborately designed features. Another challenge that some crucial information as well interactions among words or arguments may be ignored since most works treat sentences equally. Therefore, we employ Bidirectional Long Short Term...
Abstract Hepatocellular carcinoma (HCC) is a prevalent malignancy and there lack of effective biomarkers for HCC diagnosis. Living organisms are complex, different omics molecules interact with each other to implement various biological functions. Genomics metabolomics, which the top bottom systems biology, play an important role in clinical management. Fatty acid metabolism associated malignancy, prognosis, immune phenotype cancer, potential hallmark malignant tumors. In this study, genes...
We participate in the BB3 and GE4 tasks of BioNLP-ST 2016.In task, we adopt word representation methods to improve feature-based Biomedical Event Extraction System, take 4th place.In based on Uturku system, a two-stage method is proposed for trigger detection, which divides detection into recognition stage classification stage, using different features each stage.In edge Passiveaggressive (PA) online algorithm, then constitute events by post-processing TEES.
During the development of complex diseases, there is a critical transition from one status to another at tipping point, which can be an early indicator disease deterioration. To effectively enhance performance risk identification, novel dynamic network construction algorithm for identifying warning signals based on data-driven approach (EWS-DDA) was proposed. In EWS-DDA, shrunken centroid introduced measure expression changes in assumed pathway reactions during progression and define by...
Due to the vigorous development of big data, news topic text classification has received extensive attention, and accuracy semantic analysis are worth us explore. The information contained in an important impact on results. Traditional methods tend default structure sequential linear structure, then classify by giving weight words or according frequency value words, while ignoring text, which eventually leads poor In order solve above problems, this paper proposes a BiLSTM-GCN (Bidirectional...
Biomedical event detection is a pivotal information extraction task in molecular biology and biomedical research, which provides inspiration for the medical search, disease prevention, new drug development. The existing methods usually detect simple events complex with same model, performance of relatively low. In this paper, we build different neural networks respectively, helps to promote extraction. To avoid redundant information, design dynamic path planning strategy argument detection....
Spatial transcriptomics (ST) technologies have transformed our ability to study tissue architecture by capturing gene expression profiles along with their spatial context. However, the high-dimensional ST data often limited resolution and exhibit considerable noise sparsity, thus posing significant challenges for deciphering subtle patterns. To address these challenges, we introduce DeepFuseNMF, a novel multi-modal dimensionality reduction framework that enhances integrating low-resolution...