A5101772460- Caching and Content Delivery
- Advanced Text Analysis Techniques
- Peer-to-Peer Network Technologies
- Advanced Computational Techniques and Applications
- Sentiment Analysis and Opinion Mining
- Text and Document Classification Technologies
- Photopolymerization techniques and applications
- Power Systems and Technologies
- Energy Efficient Wireless Sensor Networks
- Service-Oriented Architecture and Web Services
- Conducting polymers and applications
- Topic Modeling
- Opportunistic and Delay-Tolerant Networks
- Semantic Web and Ontologies
- Web Applications and Data Management
- AI-based Problem Solving and Planning
- Mental Health via Writing
- Expert finding and Q&A systems
- Military Defense Systems Analysis
- Access Control and Trust
- Artificial Intelligence in Law
- Face and Expression Recognition
- Molecular Junctions and Nanostructures
- Generative Adversarial Networks and Image Synthesis
- Safety and Risk Management
Hubei University of Technology
2011-2024
University of Science and Technology Beijing
2022-2024
PLA Information Engineering University
2023
Liaoning Academy of Materials
2023
Chongqing Municipal Government
2015
University of Georgia
2007-2012
University of Akron
2003
Wireless sensor networks have recently gained a lot of attention. Target tracking is canonical application wireless and task allocation an unavoidable problem for target tracking. To improve the quality save more energy, auction-based dynamic coalition scheme single in was introduced. Furthermore, extended to multiple achieve even better performance. The simulations compared standard deviation error, detecting ratio energy consumption new schemes with those case-based reasoning scheme. saves...
Multispectral image reconstruction, which aims to recover a three-dimensional (3D) spatial–spectral signal from two-dimensional measurement in spectral camera based on ghost imaging via sparsity constraint (GISC), has been attracting much attention recently. However, faced with abundant 3D data, the reconstruction quality cannot meet visual requirements. Based robust data processing capability of deep learning, novel network called SSTU-Net3+ is constructed by improving U-Net3+ transformer...
Increased commodity use of mobile devices has the potential to enable mission-critical monitoring applications. However, these mobile-enabled applications have often work in environments where a delay-tolerant network (DTN) is only feasible communication paradigm. Detection complex (composite) events fundamental existing plan-based CED techniques are mostly centralized, and hence inherently unscalable for DTNs. In this paper, we create Comet â" decentralized plan-based, efficient scalable...
As the lifeblood of electric power system, fault transmission lines directly threaten safe operation system. Thus, effective and accurate prediction positioning analysis lines, has important practical value economic significance to security To solve asymmetry line problem, paper proposes a novel approach utilize fuzzy K-NN(K-Nearest Neighbor) classifier predict types position line. During training phase, proposed uses connection between data obtain relevant cluster center features greatly...
Liquid crystalline elastomers (LCEs) are excellent candidates for actuators and soft robotics owing to their exceptional properties. Their actuation relies on the precise control of liquid orientation within elastomeric network during synthesis. Current two-stage synthesis methods, which involve consecutive thiol-acrylate addition (or amine-acrylate addition), mechanical stretching, free-radical polymerization residual acrylate, typically suffer from extended reaction times oxygen inhibition...
Plan recognition is a ubiquitous task in artificial intelligence and pervasive computing research. The multigoal problem presents major challenge the real world of plan recognition. Users often pursue several goals concurrent interleaving manner, where pursuit may spread over different parts an activity sequence be pursued parallel. Existing approaches for problems are probabilistic approaches. They all assume existence libraries, which require lot human efforts predicting formalizing plans...
Abstract Relational extraction plays an important role in the field of natural language processing to predict semantic relationships between entities a sentence. Currently, most models have typically utilized tools capture high-level features with attention mechanism mitigate adverse effects noise sentences for prediction results. However, task relational classification, these mechanisms do not take full advantage information some keywords which on expressions sentences. Therefore, we...
The utility and the effectiveness of peer-to-peer (P2P) content distribution systems can be greatly enhanced by augmenting their ad-hoc discovery mechanisms with two capabilities, namely a mechanism to enable peers register queries receive notifications when corresponding data-items are added network means for advertise new content. While P2P-based publish-sub scribe infuse these developing full-fledged publish-subscribe on top unstructured P2P networks requires complex techniques, it is...
With the rapid spread of web application, search engine based on personalized services becomes more important in society. The paper proposes an approach to design and implement a Heritrix Lucene. In particular, crawler mobile information with URL hashing algorithm, achieving multithread efficient crawl. Experimental results show performance.
Many e-learning systems are proposed and used practically due to the rich opportunities provided by Internet. In particular, technology implementation in a positive learning experience can be one of strongest influencing factors stimulating learners' motivation. From perspective learning, paper designed implemented new system, which incorporated three innovative internet technologies into well-known open system named Chamilo. The were: easy knowledge search classification, instant chat...
We investigate the irreversible adsorption of poly(3-hexylthiophene) (P3HT) polymer thin films on silicon dioxide/silicon (SiO
Mining valuable services from the massive big data of court judgment texts is focus current judicial informatization research. The paper proposed a novel TF-W2V comprehensive similarity model based on deep neural network to calculate texts.The approach can implement semantic retrieval similar with cases that users want search, and recommend critical information needed by courts in process handling cases, which helpful for new case decision-making. Experimental results show accuracy up 37%...
Abstract Recently, convolutional neural networks (CNNs) have achieved excellent performance for the recommendation system by extracting deep features and building collaborative filtering models. However, CNNs been verified susceptible to adversarial examples. This is because samples are subtle non-random disturbances, which indicates that machine learning models produce incorrect outputs. Therefore, we propose a novel model of Adversarial Neural Collaborative Filtering with Embedding...
The sequential recommendation aims to capture the user's dynamic interest characteristics according historical interaction sequences of users, provide users with much more personalized recommendations. Although popular transformer-based models can effectively items dependencies between sequences, they also have some problems such as noisy information and data sparseness in reality. Therefore, this paper proposes a novel model based on Filter Model Augmentation Strategies, named FMAS short....
The publish-subscribe (pub-sub) paradigm is maturing and integrating into community-oriented collaborative applications. Because of this, pub-sub systems are faced with an event stream that may potentially contain large numbers redundant partial messages. Most view
To enhance the competitiveness of colleges and universities in graduate enrollment reduce pressure on candidates for examination consultation, it is necessary practically significant to develop an intelligent Q&A platform, which can understand analyze users' semantics accurately return information they need. However, there are problems such as low volume quality corpus enrollment, this paper develops a question answering platform based novel retrieval model including density-based...
Since the whole world has been facing unprecedented coronavirus health crisis for almost a year now, real-time management of emergencies, such as awareness prevention and control during epidemic situations, paid more attention by universities every country. Therefore, paper develops implements novel emergency system based on Apache Storm students universities, named CSEMS, which is abbreviation College Students Emergency Management System, to provide kind university support universities. To...