A5100331933- Natural Language Processing Techniques
- Semantic Web and Ontologies
- Topic Modeling
- Advanced Computational Techniques and Applications
- Advanced Vision and Imaging
- Service-Oriented Architecture and Web Services
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Video Surveillance and Tracking Methods
- Data Management and Algorithms
- Web Data Mining and Analysis
- Image Processing Techniques and Applications
- Artificial Intelligence in Healthcare
- Radiomics and Machine Learning in Medical Imaging
- Information Retrieval and Search Behavior
- Intelligent Tutoring Systems and Adaptive Learning
- Digital Media Forensic Detection
- 3D Shape Modeling and Analysis
- Imbalanced Data Classification Techniques
- Distributed Control Multi-Agent Systems
- Computational Drug Discovery Methods
- Medical Image Segmentation Techniques
- Solar Radiation and Photovoltaics
- Recommender Systems and Techniques
- Speech Recognition and Synthesis
Zhejiang Normal University
2025
Sichuan University
2023
Nankai University
2022
Sun Yat-sen University Cancer Center
2021
State Key Laboratory of Oncology in South China
2021
Institute of Automation
2021
University of Chinese Academy of Sciences
2021
Sun Yat-sen University
2021
Peking University
2021
Hunan University
2020
Tumor-infiltrating lymphocytes (TILs) are clinically significant in triple-negative breast cancer (TNBC). Although a standardized methodology for visual TILs assessment (VTA) exists, it has several inherent limitations. We established deep learning-based computational TIL (CTA) method broadly following VTA guideline and compared with TNBC to determine the prognostic value of CTA reasonable workflow clinical practice.We trained three neural networks nuclei segmentation, classification...
Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural processing tasks that were previously thought to be exclusive humans. In this work, we introduce Qwen, first installment our large model series. Qwen is a comprehensive series encompasses distinct with varying parameter counts. It includes base pretrained models, and Qwen-Chat, chat finetuned human alignment techniques. The consistently demonstrate superior performance across multitude...
Sensor Network has been widely used in all fields, and a large number of heterogeneous data is produced.In order to make these shared integrated semantically, semantic web technology introduced the sensor network.In this paper, we comprehensive survey about Semantic Web (SSW) from perspective ontology, mapping query respectively.Firstly, present state art SSW three aspects: ontologies before SSN ontology extensions which are emphatically.Secondly, discuss stream focus on comparison languages...
Identification of drug-target interactions (DTIs) is crucial for drug discovery and repositioning. Existing graph neural network (GNN) based methods only aggregate information from directly connected nodes restricted in a drug-related or target-related network, are incapable capturing long-range dependencies the biological heterogeneous graph. In this paper, we propose attention (HGAN) to capture complex structures rich semantics DTI prediction. HGAN enhances structure learning both...
Accurate ball tracking in sports is vital for automatic analysis yet it challenging mainly due to the small size and occlusions. This study proposes a novel multi‐camera 3D (MBT) framework video. The proposed consists of four parts: 2D detection, tracking, position fusion, tracking. In aspect, multi‐scale features are introduced enhance also improved by exploring cross‐view information handle occlusion timely updating model with detection results alleviate problem drift. For ball, fusion...
This paper describes a generative model for extracting medical terms and their status from Chinese dialogues. Notably, the extracted semantic information plays an essential role in downstream tasks such as automatic scribe diagnosis system. However, how to effectively leverage dialogue context generate corresponding accurately remains less explored. Existing methods treat text concentrated long without considering characteristics of conversation, colloquialism, redundancy, interactions, etc....
This paper incorporates sampling-based global path planning with model predictive image-based visual servoing (IBVS) for quadrotor unmanned aerial vehicles (UAVs) equipped a fixed camera. The proposed method produces safe control inputs of in obstacle environment by taking image feature kinematics into account. Firstly, we utilize algorithm optimal planning, i.e., rapidly-exploring random tree (RRT*), to extend search camera space iteratively and generate sequence waypoints between the...
With the character of high incidence, prevalence and mortality, stroke has brought a heavy burden to families society in China. In 2009, Ministry Health China launched national screening intervention program, which screens risk factors conducts high-risk population interventions for people aged over 40 years old all this include hypertension, diabetes, dyslipidemia, atrial fibrillation, smoking, lack exercise, apparently overweight or obese family history stroke. People with more than two...
Mingfeng Xue, Dayiheng Liu, Wenqiang Lei, Jie Fu, Jian Lan, Mei Li, Baosong Yang, Jun Xie, Yidan Zhang, Dezhong Peng, Jiancheng Lv. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.
Emotional dialogue generation aims to generate appropriate responses that are content relevant with the query and emotion consistent given tag. Previous work mainly focuses on incorporating information into sequence or conditional variational auto-encoder (CVAE) models, they usually utilize tag as a feature influence response process. However, cannot well guarantee consistency between In this article, we propose novel Dual-View CVAE model explicitly relevance jointly. These two views gather...
This paper presents a new passive image authenticate algorithm to check and measure the forged pictures images in regional copies sticks. After reducing dimension by DWT (Discrete Wavelet Transform), Tchebichef moment invariants is applied fixed sized overlapping blocks of low-frequency wavelet sub-band, eigenvectors are lexicographically sorted. Then, similar matched certain threshold. Finally, forgery part identified threshold analysis. The experimental results show that proposed method...
As a newly shopping tool, electronic commerce has been drawing more and attention of researchers. According to the characteristics comments diversity, it is necessary extract evaluation object which an important component sentiment information. This paper explores Conditional Random Field (CRF) do objects extraction. After observing generally used features in extraction, this conclude all into four categories, i.e. word Segmentation, Part-of-speech Tagging (POS), Dependency Parsing, Semantic...