A5100440519- Higher Education and Teaching Methods
- Digital Media Forensic Detection
- Advanced Steganography and Watermarking Techniques
- Image and Video Stabilization
- Topic Modeling
- Chaos-based Image/Signal Encryption
- Recommender Systems and Techniques
- Bayesian Modeling and Causal Inference
- Education and Vocational Training
- Remote-Sensing Image Classification
- Target Tracking and Data Fusion in Sensor Networks
- Soil Mechanics and Vehicle Dynamics
- Software System Performance and Reliability
- Educational Technology and Assessment
- IoT and Edge/Fog Computing
- Ideological and Political Education
- Safety and Risk Management
- Sentiment Analysis and Opinion Mining
- 3D Modeling in Geospatial Applications
- Advanced Decision-Making Techniques
- Multi-Criteria Decision Making
- Vehicle Dynamics and Control Systems
- Biomedical Text Mining and Ontologies
- Innovative Educational Techniques
- Remote Sensing and Land Use
Jilin University
2006-2024
Energy Research Institute
2019-2021
Heilongjiang Institute of Technology
2011
Jilin Medical University
2006
Jiangxi Transportation Research Institute
2005
To improve the accuracy of user implicit rating prediction, we combine traditional latent factor model (LFM) and bidirectional gated recurrent unit neural network (BiGRU) to propose a hybrid that deeply mines semantics in unstructured content text generates more accurate matrix. First, utilize user's historical behavior (favorites records) build matrix decompose obtain vectors users literature. We also apply BERT for word embedding research papers sequence vectors. Then, BiGRU with attention...
The future network slicing enabled mobile ecosystem is expected to support a wide set of heterogenous vertical services over common infrastructure. service robustness and their intrinsic requirements, together with the heterogeneity infrastructure resources in both technological spatial domain, significantly increase complexity create new challenges regarding management orchestration. High degree automation, flexibility programmability are becoming fundamental architectural features enable...
A kind of digital audio watermarking algorithm based on discrete wavelet transform (DWT) is proposed in the paper. In paper, we adopt remote sensing image as (such satellite image). order to enhance into audio, information dealt with error diffusion method which one half-toning technique. Using chaos sequence for synchronization code and embedded it water-marking, thus has ability selfsynchronization. The quantizes lowestfrequency coefficients DWT signals, so that can be detected a blinding...
Abstract: Complex system modeling by subjective judgment alone can not meet the needs of larger and more complex actual systems. The expert knowledge base in belief rule (BRB) participate well, but there are still following problems. On one hand, BRB cannot deal with explosion combination rules caused excessive number attributes. other parameters may lose their interpretability after modeling, reasoning, optimization. In view above two problems, an interval construction for systems (IBRB-i)...
With Human-Centric Research (HCR) we can steer research activities so that the outcome is beneficial for human stakeholders, such as end users. But what exactly makes human-centric? We address this question by providing a working definition and define how pipeline be split into different stages in which human-centric components added. Additionally, discuss existing NLP with HCR series of guiding questions, serve starting points researchers interested exploring approaches. hope work would...
A kind of digital audio watermarking algorithm based on discrete wavelet transform (DWT) is proposed in the paper. In paper, we adopt remote sensing image as (such satellite image). order to enhance into audio, information dealt with error diffusion method which one half-toning technique. Using chaos sequence for synchronization code and embedded it water-marking, thus has ability selfsynchronization. The quantizes lowestfrequency coefficients DWT signals, so that can be detected a blinding...
In this paper, an adaptive blind watermarking algorithm is proposed, which presents a new method of block classification based on the characteristic visual masking HVS (human system) model, including energy and variance. According to method, classify image blocks select quantization pedometers adaptively embed watermark into DC component. At same time, because shortage intensity only embedded components, that mid-frequency subbands by changing order these coefficients under invisibility....
In this paper, an adaptive blind watermarking algorithm is proposed, which presents a new method of block classification based on the characteristic visual masking HVS (human system) model, including energy and variance. According to method, classify image blocks select quantization pedometers adaptively embed watermark into DC component. At same time, because shortage intensity only embedded components, that mid-frequency subbands by changing order these coefficients under invisibility....
Graph neural networks have recently received increasing attention. These methods often map nodes into latent spaces and learn vector representations of the for a variety downstream tasks. To gain trust to promote collaboration between AIs humans, it would be better if those were interpretable humans. However, most explainable focus on supervised learning setting aim answer following question: "Why does model predict y an input x?". For unsupervised as node embedding, interpretation can more...
Higher education institutions have a moral as well legal responsibility to protect the sensitive data maintained on campus computers. are beginning look at areas improve protection of their information technology structure due increase viruses, spam, hackers, and identity theft. The main goal this paper is study network security in university. result shows genuine exposure, risk being effectively compromised real. Consequently, trap should be closely monitored control mechanisms, such...
We see widespread adoption of slate recommender systems, where an ordered item list is fed to the user based on interests and items' content. For each recommendation, can select one or several items from for further interaction. In this setting, significant impact behaviors mutual influence among well understood. The existing methods add another step re-ranking after ranking stage which considers recommended re-rank generate recommendation results so as maximize expected overall utility....
Abstract Heavy metal pollution of crops seriously endangers food security and indirectly threatens human health. Direct measures in the fields laboratories through on-site sample collection, testing, analysis are time-consuming labor intensive, thereby prohibiting their applications large-scale monitoring. Remote sensing techniques provide an alternative means examining above-ground vegetation status, e.g. leaf area index (LAI). metals, however, typically accumulated root crops, which may...
ADVERTISEMENT RETURN TO ISSUEPREVAddition/CorrectionNEXTORIGINAL ARTICLEThis notice is a correctionAnalysis of Concentration Multiplicity Patterns Continuous Isothermal Mixed Suspension−Mixed Product Removal Reactive Precipitators.Qiuxiang Yin, Jingkang Wang, Zhao Xu, and Guizhi LiCite this: Ind. Eng. Chem. Res. 2001, 40, 7, 1791Publication Date (Web):February 21, 2001Publication History Published online21 February 2001Published inissue 1 April...
The electricity transmission and distribution tariff policy of the second supervision cycle in China has formulated a much better system. In this context, research on risk related to regulation faced by power grid enterprises is helpful for regulatory agencies business operators identify avoid risks time promote sustainable development electric industry. Firstly, evaluation criteria system established. Secondly, model based best worst method (BWM) cloud proposed. Finally, level four...