A5101943274- Advanced Computational Techniques and Applications
- Educational Technology and Assessment
- Software System Performance and Reliability
- Advanced Decision-Making Techniques
- Software Testing and Debugging Techniques
- Service-Oriented Architecture and Web Services
- Higher Education and Teaching Methods
- Software Reliability and Analysis Research
- Renal Transplantation Outcomes and Treatments
- Organ Transplantation Techniques and Outcomes
- Engineering and Test Systems
- Organ Donation and Transplantation
- Web Data Mining and Analysis
- Online Learning and Analytics
- Power Systems and Technologies
- Text and Document Classification Technologies
- Imbalanced Data Classification Techniques
- Software Engineering Research
- Artificial Immune Systems Applications
- Sustainable Industrial Ecology
- Advanced Steganography and Watermarking Techniques
- Industrial Vision Systems and Defect Detection
- Sentiment Analysis and Opinion Mining
- Advanced Graph Neural Networks
- Innovative Educational Techniques
Taishan University
2022-2025
Tianjin Normal University
2021-2024
Chinese Academy of Agricultural Engineering
2023
Ministry of Agriculture and Rural Affairs
2023
Southwest University
2008-2021
First Affiliated Hospital of Xi'an Jiaotong University
2018-2020
Nanyang Polytechnic
2020
Xi'an Jiaotong University
2020
Shandong University of Science and Technology
2012
Beijing Jiaotong University
2006-2010
Chronic Kidney disease is a severe lifelong condition caused either by renal or impaired functions of the kidneys. In present area research, cancer one deadliest and crucial importance for survival patients ' diagnosis classification. Early proper therapy can stop delay development this chronic into final stage where dialysis transplantation only way saving life patient. The automated tools to accurately identify subtypes kidney is, therefore, an urgent challenge in recent past. paper,...
The evolution of generative artificial intelligence technology offers new possibili-ties for the digital transformation higher education classrooms. This study takes domestic intelligent model DeepSeek as research object to explore its functional positioning, application value, and potential challenges in teaching scenarios colleges universities. Through literature analysis theoretical construction, combined with observations practices, it proposes that assumes three roles classroom...
Reducing the dimensions of high-dimensional feature set is one difficulties text categorization. Feature selection has been effectively applied in classification, because its low complexity computing. Research works show that mutual information a good method but doesn't consider term frequency each category corpus and connections between terms. To remedying defects traditional method, this article improved measure by introducing class dispersion class, built experimental platform...
This article is based on the ecological security of Hadoop cluster.Combined with practical application Kerberos authentication system in cluster, we propose a multi node HA (High Available) configuration Kerberos, realize Master-Master-Slave design mode and increase redundant nodes storage.Based characteristics dual Master, set timed tasks, execute scripts time, carry out data dump Master nodes, prevent single failure, avoid failure system.We use Slave to one way backup storage from...
There are many problems in traditional network security and PC protection systems, such as insecurity, inflexibility so on, whether they implemented by software or hardware. To solve the problems, this paper presents a reconfigurable system on basis of NetFPGA platform embedded soft-core technology. This consists two subsystems, one is for subnet other DE2 board to protect terminal, which used achieve packet filtering hardware, immunity from ARP attacks flow monitoring transmitting with...
This paper shares our work in developing and implementing an immersive gamification training platform for students who undergo manufacturing shopfloor at the School of Engineering, Nanyang Polytechnic, Singapore. In this platform, we developed a virtual that is identical to actual located school. Students have freedom learn operations safety acts through various game scenarios tasks which include workshop safety, CNC machine introduction, machining dynamics, MES, etc. addition, assessment...
Abstract In this paper, the characteristics of software defect prediction are analyzed from perspective machine learning. To solve problem some redundant or uncorrelated features in data sets, a stable feature selection method based on relevancy and redundancy (RRSFS) is proposed. RRSFS combines between correlation classes to select optimal subset. not only reduces cost operation model, but also enhances stability algorithm.
This paper proposes a promising palmprint image enhancement algorithm. Under the constraint that keeps original characteristic of palmprint, method improves contrast images, in order to extract features from by Scale Invariance Feature Transformation (SIFT) descriptor. However, existing algorithms can hardly yield SIFT feature low resolution e.g., dpi these images is less than 150. Our scheme uses local entropy reassign coefficients traditional Unsharp Mask (UM) algorithm, for an...
This paper examines and discusses the course of C language program design in accordance with requirements training applied talents background new engineering courses, it presents particular answers to issues that arise instruction. Putting student at center helps create a mixed teaching pattern, curriculum designed for program, specialized elective course, incorporates politics into through project practice fusion. investigates reform strategies approaches against backdrop "new course" by...
In view of the shortcomings test data generation algorithm including particle swarm optimization and ant colony algorithm, a new is proposed, which based on combination parameter adjustment.This can dynamically adjust its search capabilities fitness value particles , combine advantages (PSO) algorithm(ACO)to ensure convergence accuracy algorithm.Experiments show that effectively improve efficiency generation.
Based on user's browsing path, a novel clustering model is proposed to cluster Web users. It represents users' characteristics by their transaction paths and presents new similarity computational method-WUSC (Web user calculation). Considering the relationship between common path whole of user, combining with residence time calculate users, are represented as sequential patterns,. The experiment results prove efficiency method.
Abstract This paper mainly analyzes the characteristics of software defect prediction from perspective machine learning, and proposes a semi-supervised method based on sampling integration for problem class imbalance in data incomplete classification sets. SISDP). SISDP firstly constructs robust KNN marking model by taking balanced sample samples to mark batch unmarked data, then iteratively adds newly marked original set next model. , iterate until is marked. For set, hybrid algorithm used...