A5100765135- Visual Attention and Saliency Detection
- Digital Media Forensic Detection
- Advanced Optical Network Technologies
- Infrared Target Detection Methodologies
- Stock Market Forecasting Methods
- Satellite Communication Systems
- Network Security and Intrusion Detection
- Cloud Data Security Solutions
- Noise Effects and Management
- Technology and Data Analysis
- Face recognition and analysis
- Advanced Data Storage Technologies
- Customer churn and segmentation
- Evaluation Methods in Various Fields
- Internet Traffic Analysis and Secure E-voting
- Software-Defined Networks and 5G
- Olfactory and Sensory Function Studies
- Network Packet Processing and Optimization
- Biometric Identification and Security
- Generative Adversarial Networks and Image Synthesis
- Anomaly Detection Techniques and Applications
- Recommender Systems and Techniques
- Visual perception and processing mechanisms
- IoT-based Smart Home Systems
- Context-Aware Activity Recognition Systems
Lanzhou University
2021
Beijing University of Posts and Telecommunications
2014-2020
London South Bank University
2015
Jilin Weather Modification Office
2015
Integrated Space-Terrestrial Network (ISTN) can effectively improve the global communication coverage and SDN framework in ISTN has emerged as a promising solution to guarantee Quality of Service (QoS). At present, researches on routing algorithm are mainly focused QoS parameters about resource availability do not consider impact network attacks comprehensively, such DDoS attack. Plenty available resources will be occupied by normal traffic may blocked incorrectly attack defense solutions....
Predicting consumer profitability dynamically over time plays a vital role in today's customer-centric business. In this paper, we adopt dynamic systems approach to address the prediction problem of customer profitability. Based on transaction records, RFM score-based series are generated using cluster analysis. These used measure and describe Furthermore, multilayer feed-forward neural network models trained capture dynamics evolving A set real transactions from UK-based online retailer is...
Wireless smart home system is to facilitate people's lives and it trends adopt a more intelligent way provide services. It beneficial design an (SH) precisely recognize users' behaviors automatically responsd with the corresponding activities satisfy actual demands. However, activity models in existing approaches are usually constructed separately through statistic probability. These cannot user's dynamical intentions accurately. To address problem, we propose new SH architecture device...
In traditional prediction methods, we have to overcome the errors when predict for dynamic data. this paper, theory and moving average forecasting has been combined. We had designed a kind of weighted method data time series. Dynamic weighting parameter constructed in original forecast. The effect is better than forecast compared relative error. At last, predicted Urumqi precipitation 30 years with method. error smaller.
With the development of Internet Things (IoT), people's lives have become increasingly convenient. It is desirable for smart home (SH) systems to integrate and leverage enormous information available from IoT. Information can be analyzed learn user intentions automatically provide appropriate services. However, existing service recommendation models typically do not consider services that are unavailable in a user's living environment. In order address this problem, we propose series...
The emergence of integrated space-ground network (ISGN), with more complex conditions compared tradition network, requires packet classification to achieve high performance. Packet plays an important role in the field security. Although several existing schemes have been proposed recently improve performance, performance these is unable meet high-speed requirement ISGN. To tackle this problem, a hybrid algorithm based on hash table and geometric space partition (HGSP) paper. HGSP falls into...
In recent years, deep learning has entailed great progresses, but this success is seen as a triumph of empiricism. Despite the efforts made in understanding models, gap between theoretical interpretability and empirical models still large. At present, analysis mostly done after training process. work, novel method proposed to extract interpretable features during Particularly, modified Siamese network used disentangle using specially designed loss function. After feature extractor, fully...