A5016193325- Topic Modeling
- Corporate Finance and Governance
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Data Management and Algorithms
- Advanced Computational Techniques and Applications
- Natural Language Processing Techniques
- Advanced Algorithms and Applications
- Corporate Social Responsibility Reporting
- Text and Document Classification Technologies
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Image Retrieval and Classification Techniques
- Data Stream Mining Techniques
- Auditing, Earnings Management, Governance
- Web Data Mining and Analysis
- Digital Marketing and Social Media
- Opinion Dynamics and Social Influence
- Human Mobility and Location-Based Analysis
- Multimodal Machine Learning Applications
- Traffic Prediction and Management Techniques
- Complex Systems and Time Series Analysis
- Capital Investment and Risk Analysis
- Fuzzy Logic and Control Systems
- Machine Learning and Algorithms
Zhongnan University of Economics and Law
2015-2024
Chinese Academy of Sciences
2024
Shanghai University of Electric Power
2024
Institute of Computing Technology
2024
Zhengzhou University
2024
PLA Army Engineering University
2010-2023
Tangshan College
2012-2018
Illinois Institute of Technology
2018
United States Army Corps of Engineers
2011-2012
Chinese Academy of Medical Sciences & Peking Union Medical College
2012
This study proposes a short-term traffic flow prediction model that combines community detection-based federated learning with graph convolutional network (GCN) to alleviate the time-consuming training, higher communication costs, and data privacy risks of global GCNs as amount increases. The GCN (FCGCN) can achieve timely, accurate, safe state predictions in era big data, which is critical for efficient operation intelligent transportation systems. FCGCN process has four steps: dividing...
Deep generative models have gained much attention given their ability to generate data for applications as varied healthcare financial technology surveillance, and many more - the most popular being adversarial networks (GANs) variational auto-encoders (VAEs). Yet, with all machine learning models, ever is concern over security breaches privacy leaks deep are no exception. In fact, these advanced so rapidly in recent years that work on still its infancy. an attempt audit current future...
In this work, a ternary BiOCl/g-C3N4/Ag2CrO4 composite with dual Z-scheme heterojunction was prepared based on g-C3N4 photocatalyst and applied in the photocatalytic reduction of CO2. The optimal CH4 CO yields catalyst could be achieved BOC/CN/ACO-1% 81.21 30.20 μmol g–1 for 6 h, respectively, which considerably enhanced compared to that pure g-C3N4. apparent quantum efficiency 2.68% photoreduction CO2 at 420 nm. This performance enhancement resulted from constructing improved light...
This study draw upon the theory of habit formation in consumption from macroeconomics to support evidence on existence social media consumption. Treating as a form digital good and using aggregated weekly posts Facebook pages group 12 politicians cabinet Singapore, we verified through non-separable recursive time model that habits were developed among this politicians. further confirms reciprocity by validating citizens followers these politicians' data 'likes', 'shares' 'comments'. Further,...
Named entity recognition (NER) is an indispensable and very important part of many natural language processing technologies, such as information extraction, retrieval, intelligent Q & A. This paper describes the development AL-CRF model, which a NER approach based on active learning (AL). The algorithmic sequence processes performed by model following: first, samples are clustered using k -means approach. Then, stratified sampling produced clusters in order to obtain initial samples,...
There are not many real-time public mood tracking frameworks over social media streams at present. Real-time microblogs becomes necessary for further studies with low-latency requirements. To address this issue, we propose a hierarchical framework time series Chinese microblog using complex event processing. Complex processing is able to handle high-speed and high-volume data streams. First, transform into emotional events through the text sentiment analysis. Then, apply an online batch...
The purpose of this study is to apply noncoplanar intensity‐modulated radiation therapy (Nonco_IMRT) young female patients with mediastinal lymphoma. Nonco_IMRT was evaluated through a planning comparison coplanar IMRT (Co_IMRT) and conventional anteroposterior posteroanterior fields (AP–PA) plans. Co_IMRT performed five equally spaced beams starting from gantry angle . used two in the sagittal plane replace that directly irradiated breasts. Nineteen were enrolled retrospective study. Dose...
The Particle Swarm Optimization (PSO) is a heuristic optimization technique-based swarm intelligence that can be applied to solving many real-world problems. However, the standard PSO algorithm easily get trapped in local optima and has slow convergence speed, these drawbac ks have hindered its further development all fields. In this paper, new method based on neighbor Gaussian cloud learning introduced order improve performance of traditional (NHPSO). NHPSO consists two main steps. First,...
Purpose – This study aims to build on the organizational learning theory and propose a complex strategy by combining strategic alliance with subsequent acquisitions penetrate new product markets. The authors empirically examined whether what extent preacquisition experience affects short- long-term stock performance of acquiring firms. Design/methodology/approach Data acquisitions, in which acquirers have from activities their targets’ respective industry, were collected. Diversifying...