A5101731642- Sentiment Analysis and Opinion Mining
- Botulinum Toxin and Related Neurological Disorders
- Topic Modeling
- Neurological disorders and treatments
- Advanced Text Analysis Techniques
- Cooperative Communication and Network Coding
- Emotion and Mood Recognition
- Error Correcting Code Techniques
- IoT and Edge/Fog Computing
- Hand Gesture Recognition Systems
- Software System Performance and Reliability
- Network Security and Intrusion Detection
- Genetic Neurodegenerative Diseases
- Speech and Audio Processing
- Text and Document Classification Technologies
- Advanced Malware Detection Techniques
- Cloud Computing and Resource Management
- Software-Defined Networks and 5G
- Coding theory and cryptography
- Natural Language Processing Techniques
- Caching and Content Delivery
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Phonocardiography and Auscultation Techniques
- Image Retrieval and Classification Techniques
Ningbo Medical Center Lihuili Hospital
2025
Ningbo University
2025
Zhejiang University of Technology
2022-2023
Northeast Electric Power University
2022-2023
Electric Power University
2022
University of California System
2022
Tongji University
2012-2021
Tongji Hospital
2019-2021
Yulin University
2021
Southern University of Science and Technology
2020
Third-party libraries (TPLs) have become a significant part of the Android ecosystem. Developers can employ various TPLs with different functionalities to facilitate their app development. Unfortunately, popularity also brings new challenges and even threats. may carry malicious or vulnerable code, which infect popular apps pose threats mobile users. Besides, code third-party could constitute noises in some downstream tasks (e.g., malware repackaged detection). Thus, researchers developed...
In this paper, we use a wireless sensor node to investigate the relationship between vocal art composition and human body noise. For music, suggested predictor created convergence link model HART graph routing is used systematically design noise science. The optimizes acquisition of information on analyzes acoustic signal characteristics. optimally balanced scheduling has been scientifically distributedly regulate performance proposed enhances detection ability raises level art, music...
Electrical power supply of data centers constitutes major operational costs. In this paper, we focus on providing the supplying energy to exact match consumption demands by using properties a digital grid. Different from present grid, where is supplied continuously and in amounts satisfy large demands, grid supplies form packets upon request. Energy are specifically addressed one, or multiple servers no others, they carry amount that has been requested. To evaluate level supply, model time...
1Finding persistent and inactive activity periods is very helpful in practice, for example to detect intrusion activities. Most of the literature focuses on finding flows or frequent flows. No previous work able find infrequent In this paper, we propose a novel sketch data structure, PISketch, real time. The key idea PISketch define weight its Reward Penalty System each flow combine balance information both persistency infrequency, keep high-weighted limited space through strategy. We...
The relationship between brain abnormalities and phenotypic characteristics in cervical dystonia (CD) patients has not been fully established, little is known about the neuroplastic changes induced by botulinum toxin type A (BoNT-A) treatment.Ninety-two CD presenting with rotational torticollis 45 healthy controls from our database were retrospectively screened. After clinical assessment, 92 underwent baseline magnetic resonance imaging (MRI) followed a single-dose injection of BoNT-A. Four...
This demo delivers a novel retrieval system which meets users' multi-dimensional requirements in clothing image search. In this system, users are able to use both and keywords as query inputs. We employ the color, texture, shape attributes additional descriptors further refine requirements. propose Hybrid Topic (HT) model, probabilistic network integrating multi-channel into unified framework, learn intricate semantic representation of above. The proposed model provides an effective...
Although single-photon emission computed tomography (SPECT/CT) could help to predetermine dystonic muscles in patients with cervical dystonia (CD), its efficacy aiding botulinum toxin injection is undetermined. This randomized, double-blinded study aimed assess the of SPECT/CT aided CD.Patients were randomized into group (candidate selected by and clinical evaluation) or control (clinical evaluation). Follow-ups done at two weeks (T1), one (T2), three (T3) six months (T4). The primary...
Current data centers often employ SDNs (Software-Defined Networks) that are layered in most cases. However, information overwhelming has been a limitation to SDN deployment. We present management model for center networks. In this approach, regional networks on lower layers will be aggregated and viewed as single switches upper layers. Management divided into three parts, which can seen by network managers, controllers tenants, respectively.
Sentiment-controlled text generation aims to generate texts according the given sentiment. However, most of existing studies focus only on document- or sentence-level sentiment control, leaving a gap for finer-grained control over content generated results. Fine-grained allows review express different opinions toward multiple aspects. Some previous works attempted reviews conditioned aspect-level sentiments, but they usually suffer from low adaptability and lack an annotated dataset. To...
In this paper, we present a new method based on novel 3D cross model for detecting human eye blinks. Firstly, an initial frontal face is given to constructe the model, which represents head. by using of optical flow, can track head motion in real-time without effects self-occlusion and large motion. According horizontal information position eyes be located dynamically accurately each frame. number pixels whose gray values less than certain threshold calculated when open close respectively....
This paper presents a new method for 3D head pose recovery from video by using novel cross model to track the motion. Given an initial frontal face of head, biologically-plausible is constructed represent which robust lighting and facial expression changes. By updating templates dynamically, effects self-occlusion, large motion gradual illumination changes can be diminished. When hidden or moves out scenes, reinitialized automatically. Experimental results show that efficiently recover...
Large Language Models (LLMs) have significantly advanced the field of information retrieval, particularly for reranking. Listwise LLM rerankers showcased superior performance and generalizability compared to existing supervised approaches. However, conventional listwise reranking methods lack efficiency as they provide ranking output in form a generated ordered sequence candidate passage identifiers. Further, are trained with typical language modeling objective, which treats all errors...
In the AIOps (Artificial Intelligence for IT Operations) era, accurately forecasting system states is crucial. microservices systems, this task encounters challenge of dynamic and complex spatio-temporal relationships among microservice instances, primarily due to deployments, diverse call paths, cascading effects instances. Current time-series methods, which focus mainly on intrinsic patterns, are insufficient in environments where spatial critical. Similarly, graph approaches often neglect...
This paper studies the throughput optimization for buffer-limited mobile ad hoc networks (MANETs) based on resource allocation policies. Specifically, we consider a practical MANET configuration where total limited buffer space of network node can be flexibly allocated to its source and relay buffer, two-hop scheme with control parameter α is adopted packet delivery. For such any specific setting, i.e., are fixed, first develop theoretical framework characterize dynamics derive exact...