A5034901082- Software Engineering Research
- Service-Oriented Architecture and Web Services
- Cloud Computing and Resource Management
- IoT and Edge/Fog Computing
- Software System Performance and Reliability
- Business Process Modeling and Analysis
- Software Reliability and Analysis Research
- Advanced Software Engineering Methodologies
- Soil, Finite Element Methods
- Machine Learning in Bioinformatics
- Software Testing and Debugging Techniques
- Topic Modeling
- RNA and protein synthesis mechanisms
- Environmental and Agricultural Sciences
- Protein Structure and Dynamics
- Distributed and Parallel Computing Systems
- Cloud Data Security Solutions
- Petri Nets in System Modeling
- Natural Language Processing Techniques
- Soil and Unsaturated Flow
- Software-Defined Networks and 5G
- Caching and Content Delivery
- Real-Time Systems Scheduling
- Advanced Malware Detection Techniques
- Software Engineering Techniques and Practices
East China University of Science and Technology
2016-2025
Shanghai Advanced Research Institute
2024
Shanghai Jiao Tong University
2024
Institute of Software
2023
Taiyuan University of Technology
2001-2022
Shanghai Institute of Technology
2018
China Guangzhou Analysis and Testing Center
2012-2016
Inner Mongolia Agricultural University
2000-2014
Nanjing University
2011-2012
Taiyuan University of Science and Technology
2011
In order to improve software reliability, defect prediction is applied the process of maintenance identify potential bugs. Traditional methods mainly focus on designing static code metrics, which are input into machine learning classifiers predict probabilities code. However, characteristics these artificial metrics do not contain syntactic structures and semantic information programs. Such more significant than manual can provide a accurate predictive model. this paper, we propose framework...
Deep learning has been widely used for protein engineering. However, it is limited by the lack of sufficient experimental data to train an accurate model predicting functional fitness high-order mutants. Here, we develop SESNet, a supervised deep-learning predict mutants leveraging both sequence and structure information, exploiting attention mechanism. Our integrates local evolutionary context from homologous sequences, global encoding rich semantic universal space information accounting...
Protein engineering faces challenges in finding optimal mutants from a massive pool of candidate mutants. In this study, we introduce deep-learning-based data-efficient fitness prediction tool to steer protein engineering. Our methodology establishes lightweight graph neural network scheme for structures, which efficiently analyzes the microenvironment amino acids wild-type proteins and reconstructs distribution acid sequences that are more likely pass natural selection. This serves as...
Designing protein mutants with both high stability and activity is a critical yet challenging task in engineering. Here, we introduce PRIME, deep learning model, which can suggest improved without any prior experimental mutagenesis data for the specified protein. Leveraging temperature-aware language modeling, PRIME demonstrated superior predictive ability compared to current state-of-the-art models on public dataset across 283 assays. Furthermore, validated PRIME’s predictions five...
Vehicular Ad hoc Network (VANET) is a new technology that integrates the potentials of next generation wireless networks into vehicles. The design routing protocols in VANETs crucial supporting Intelligent Transportation Systems (ITS). Typical geographic routings only use local information to make decisions which may lead maximum and sparse connectivity problems. This paper proposes novel SDN-based (SDGR) protocol for VANET, based on node location, vehicles density digital map....
Protein language models (PLMs) play a dominant role in protein representation learning. Most existing PLMs regard proteins as sequences of 20 natural amino acids. The problem with this method is that it simply divides the sequence into individual acids, ignoring fact certain residues often occur together. Therefore, inappropriate to view acids isolated tokens. Instead, should recognize frequently occurring combinations single token. In study, we use byte-pair-encoding algorithm and unigram...
Fog computing as an extension of the cloud based infrastructure, provides a better platform than for mobile computing, Internet Things, etc. One problems is how to make full use resources fog so that more requests applications can be executed on edge, reducing pressure network and ensuring time requirement tasks. The high mobility nodes also has great impact task completion user satisfaction. Thus, general IoT-Fog-Cloud architecture with contract-based resource sharing mechanism proposed in...
Fine-tuning pretrained protein language models (PLMs) has emerged as a prominent strategy for enhancing downstream prediction tasks, often outperforming traditional supervised learning approaches. As widely applied powerful technique in natural processing, employing parameter-efficient fine-tuning techniques could potentially enhance the performance of PLMs. However, direct transfer to life science tasks is nontrivial due different training strategies and data forms. To address this gap, we...
Vehicular ad hoc network (VANET) is an emerging technology for the future intelligent transportation systems (ITSs). The current researches are intensely focusing on problems of routing protocol reliability and scalability across urban VANETs. Vehicle clustering testified to be a promising approach improve by grouping vehicles together serve as foundation ITS applications. However, some prominent characteristics, like high mobility uneven spatial distribution vehicles, may affect...
Edge computing provides physical resources closer to end users, becoming a good complement cloud computing. With the rapid development of container technology and microservice architecture, orchestration has become hot issue. However, container-based scheduling problem in edge is still urgent be solved. In this paper, we first formulate as multi-objective optimization problem, aiming optimize network latency among microservices, reliability applications load balancing cluster. We further...
In recent years, drug design has been revolutionized by the application of deep learning techniques, and molecule generation is a crucial aspect this transformation. However, most current approaches do not explicitly consider apply scaffold hopping strategy when performing molecular generation. work, we propose ScaffoldGVAE, variational autoencoder based on multi-view graph neural networks, for molecules. The model integrates several important components, such as node-central edge-central...
Microservices are becoming increasingly popular in the construction of cloud applications. On basis containers, microservice instances can be implemented with high scalability and maintainability. Due to need ensuring various quality service (QoS) requirements two-layer resource structure containers virtual machines (VMs), workflow scheduling clouds is a challenging problem address. This paper proposes heuristic algorithm GSMS minimize execution cost microservice-based application while...
Cloud computing has attracted much interest recently from both industry and academia. However, the scale highly dynamic nature of cloud application imposes significant new challenges to resource management, efficient scheduling schemes are demanded. In this paper, we propose a systematic method address reliability, running time, failure processing in computing. A reflection mechanism is used abstract process as metaobject. Petri nets construct base layer model, meta metaobject protocol,...
Protein engineering is a pivotal aspect of synthetic biology, involving the modification amino acids within existing protein sequences to achieve novel or enhanced functionalities and physical properties. Accurate prediction variant effects requires thorough understanding sequence, structure, function. Deep learning methods have demonstrated remarkable performance in guiding for improved functionality. However, approaches predominantly rely on sequences, which face challenges efficiently...
Software defect prediction, which locates defective code snippets, can assist developers in finding potential bugs and assigning their testing efforts. Traditional prediction features are static metrics, only contain statistic information of programs fail to capture semantics programs, leading the degradation performance. To take full advantage metrics we propose a framework called Defect Prediction via Attention Mechanism (DP-AM) this paper. Specifically, DPAM first extracts vectors then...
Cyber-physical system (CPS) is the fuse of cyber world and dynamic physical it being widely used in areas closely related to people's livelihood. Therefore, security issues CPS have drawn a global attention an appropriate risk assessment for urgent need. The existing proposals using attack trees mainly focus on depicting possible intrusions, not interactions between threats defenses. In this paper, idea cyber-physical with use attack-defense tree (ADTree) proposed, considering effect both...
The geographically dispersed resources and ever-changing context incur unique heterogeneity, potential fragility, vulnerability of an edge-cloud system. Thus, the reliability guarantee services in is critical. This paper firstly proposes a QoS-aware scheduling model with fault-tolerance edge-cloud, which extends traditional primary-backup (PB) fault-tolerant to improve service time constraints tasks being satisfied. Then, algorithm including primary copy placement, backup placement...
Code summarization aims to generate code summaries automatically, and has attracted a lot of research interest lately. Recent approaches it commonly adopt neural machine translation techniques, which train Seq2Seq model on large corpus assume could work various new snippets. However, codes are highly varied in practice due different domains, businesses or programming styles. Therefore, is challenging learn such variety patterns into single model. In this paper, we propose brand-new framework...