A5101439062- Access Control and Trust
- Topic Modeling
- Advanced Text Analysis Techniques
- Advanced Computational Techniques and Applications
- Cloud Data Security Solutions
- Complex Network Analysis Techniques
- Text and Document Classification Technologies
- Network Security and Intrusion Detection
- Semantic Web and Ontologies
- Web Data Mining and Analysis
- Privacy-Preserving Technologies in Data
- Natural Language Processing Techniques
- Advanced Database Systems and Queries
- Cryptography and Data Security
- Security and Verification in Computing
- Privacy, Security, and Data Protection
- Service-Oriented Architecture and Web Services
- Sentiment Analysis and Opinion Mining
- Educational Technology and Pedagogy
- Law, logistics, and international trade
- Higher Education and Teaching Methods
- Infrastructure Maintenance and Monitoring
- Energy Efficient Wireless Sensor Networks
- Spam and Phishing Detection
- Advanced Image and Video Retrieval Techniques
Liaocheng University
2025
Cell Technology (China)
2025
Shanghai Key Laboratory of Trustworthy Computing
2012-2024
Anhui Medical University
2024
Dalian Maritime University
2006-2024
Hainan Normal University
2024
Beijing University of Posts and Telecommunications
2004-2023
University of Shanghai for Science and Technology
2023
Becton Dickinson (United States)
2023
University of Jinan
2022-2023
With the rapid development of Internet technology and social networks, a large number comment texts are generated on Web. In era big data, mining emotional tendency comments through artificial intelligence is helpful for timely understanding network public opinion. The sentiment analysis part intelligence, its research very meaningful obtaining trend comments. essence text classification task, different words have contributions to classification. current studies, distributed word...
PM2.5 is one of the most important pollutants related to air quality, and increase its concentration will aggravate threat people's health. Therefore, prediction surface great significance human health protection. In this study, A hybrid CNN-LSTM model developed by combining convolutional neural network (CNN) with long short-term memory (LSTM) for forecasting next 24h in Beijing, which makes full use their advantages that CNN can effectively extract features quality LSTM reflect term...
The method of text sentiment analysis based on dictionary often has the problems that doesn't contain enough words or omits some field words. In addition, due to existence polysemic with positivity, negativity, and neutrality, words' polarity cannot be accurately expressed, so accuracy is reduced extent. this paper, an extended constructed. contains basic words, which improves analysis. naive Bayesian classifier used determine in word is. Thus, value obtained. By utilizing designed score...
Abstract Zn‐I 2 battery with four‐electron reaction path (I − /I + ) in the cathode delivers high energy density, which however is thermodynamically not favored as I metastable. Herein, it demonstrated that conjugated 2P valence electrons graphitic framework can be relocated, offering chances to stabilize species. Combinations of elements (B, N, C, O) various configurations are first screened computationally, identifying O─B─C─N optimal structure. In this B‐centered domain, adjacent O and...
The mining of network sensitive information is great significance for understanding the social stability network. Obtaining public opinion helpful to master Internet users' attitudes toward important events. related artificial intelligence technology can achieve topics from texts. At present, current topic recognition model has a low rate and usually generates some inaccurate keywords. In this paper, method based on word weighted-latent Dirichlet allocation (LDA) proposed. First, basic...
In order to solve the problems of low recognition rate and less gesture categories caused by incomplete artificial feature extraction information in traditional static methods, a deep CNN framework is designed using principle convolution neural network (Convolutional Neural Network, CNN) recognize movements. Combined with variety optimal structures learning, model independent function realized. The method can not only ensure high accuracy robustness results, but also achieve speed smooth recognition.
Ransomware is a type of malware that encrypts data or locks device to extort ransom. Recently, variety high-profile ransomware attacks have been reported, and many defense systems proposed. However, none specializes in resisting untargeted such as those by remote desktop protocol (RDP) attack ransomware. To resolve this problem, paper proposes way combat RDP trapping tracing. It discovers ensnares the attacker through network deception environment uses an auxiliary tracing technology find...
With the rapid development of Internet, number Internet users has grown rapidly, and become more influential on people's lives. As a result, amount network text is increasing it difficult to extract interested event information from only by manual reading. Therefore, extraction technique automatically extracting useful large unstructured texts becomes increasingly important. Event detection first step task plays vital role in it. However, current research lacks comprehensive consideration...
Enamel has good optical and mechanical properties because of its multiscale hierarchical structure. Biomimetic construction enamel-like 3D bulk materials at nano-, micro-, mesh- macro-levels is a challenge. A novel facile, cost-effective, easy large-scale bottom-up assembly strategy to align 1D hydroxyapatite (HA) nanowires bundles enamel structure with the layer-by-layer interweaving orientation, reported. In strategy, surface oleate templated ultralong HA large aspect ratio functionalized...
To solve the problem of low diagnosis rate early weak faults rolling bearings, a novel bearing fault method based on Variational Mode Decomposition (VMD) and convolutional neural network (CNN)−Bidirectional Long Short-Term Memory (BiLSTM) was proposed. Based basic Sparrow Search Algorithm, tent chaotic mapping, Osprey Optimization Cauchy mutation were used to enhance global search ability algorithm. improve accuracy diagnosis, BiLSTM layer is introduced into CNN preserve local features...
Privacy protection is a very important issue and solutions must be developed for wide acceptance of location based services (LBSs) in wireless applications. Current approaches rely mostly on the use privacy policies to describe solve this problem an ad hoc way. In paper, we propose method three-dimensional access control model support requirements. We show that our can better user requirements LBSs. By introducing notions privacy-concerning subjects enhancing traditional mechanism,...
Cloud computing requires that customers trust a service provider’s platforms are secured and provide sufficient level of integrity for the client’s data. However, lack between cloud providers has hindered universal acceptance clouds as an increasingly popular approach processing large data sets computationally expensive programs. Existing techniques do not satisfactorily solve problem, good solution is to leverage management technology build into utility computing. Recently, modeling paid...
This work proposed a semi-supervised method for automatic texture filtering. Our leveraged limited amount of labeled data and large unlabeled to train Generative Adversarial Networks (GANs). Separate loss functions were designed both datasets. main contribution is the introduction knowledge extracted from shallow deep layers in neural networks. Loss defined within preserves edge, while identifies semantic content conversely removes small-scale variations. directly addresses major challenge...
Point cloud data with high accuracy and density is an important source for the depiction of real ground objects, there a broad research prospect using point directly 3D object detection recognition deep learning methods. However, many models in previous ignored structure information sampling randomness. To overcome this limitation, we proposed innovative model, namely, minimum bounding box over-segmentation–graph convolution network model (MBBOS-GCN) enhancing structural perception...
Privacy is one of the most important issues in providing high-quality ubiquitous network services to users over Internet. Although several privacy aware access control models have been proposed recent years, these still rely mostly on traditional designed primarily for security. When becomes a main concern, are no longer adequate. In this paper, we propose three-dimensional model enhanced with and show that our can be used as general method defining requirements systems respect protect user...
With the development of web services application, some issues security are increasingly prominent. XML is widely used its high expansibility as a platform-independent language. After analysis traditional technology, this paper formulates signature, basis and describes how to create verify signature. An implementation signature in Web illustrated Java.
A knowledge-based database assistant (KDA) which integrates a natural language query system with skeleton-based guiding facility is provided. When user works the KDA system, can supply several kinds of skeletons to guide users in performing retrieval tasks. semantic network model, S-Net, introduced represent knowledge for processing and skeleton generation. Methods implementing are discussed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Privacy is considered to be a critical issue for providing high quality ubiquitous network services users over the Internet. User's privacy should protected and access information must controlled in accordance with user's preferences. Existing privacy-aware control strategies often store all policies on server side thus fail consider dynamic nature of In this paper, we propose user-centric privacy-enhanced model show that our takes preferences into consideration can fulfill any kind...
At present, with the rapid development of scientific research, big data is not limited to economic society, but also change people's work and lifestyle at an alarming rate along, such as smart city, campus, recommend system are product data.Also, because data, there unprecedented profound changes having taken place in field education.In future, students can access all high quality education resources from cloud, which based entirely on individual needs themselves, online platform, teaching...
Chinese culture is an important force in the construction of "The Belt and Road Initiative". But as a result aging, backward means repair other reasons, ancient books are facing danger destruction. This paper intends to take literature ethnic minorities Yunnan example, aiming do virtual digital protection by introducing computer graphics, image processing, reality, somatosensory recognition new technologies field information. Firstly, minority documents collected high definition photography...