A5101904019- Higher Education and Teaching Methods
- Topic Modeling
- Natural Language Processing Techniques
- Advanced Text Analysis Techniques
- Educational Technology and Assessment
- Sentiment Analysis and Opinion Mining
- Education and Work Dynamics
- Advanced Computational Techniques and Applications
- Text and Document Classification Technologies
- Digital Media and Visual Art
- Network Traffic and Congestion Control
- Semantic Web and Ontologies
- Mobile Ad Hoc Networks
- Advanced Decision-Making Techniques
- Educational Technology and Pedagogy
- Music and Audio Processing
- Advanced Graph Neural Networks
- Data Management and Algorithms
- Software Engineering Research
- Innovative Educational Techniques
- Wireless Networks and Protocols
- Complex Network Analysis Techniques
- Advanced Database Systems and Queries
- Online Learning and Analytics
- Web and Library Services
Guangdong University of Foreign Studies
2015-2024
Hunan City University
2024
Northwest Institute of Eco-Environment and Resources
2024
Chinese Academy of Sciences
2024
Wuhan Business University
2024
Guangxi Medical University
2023
New York University
2023
First Affiliated Hospital of GuangXi Medical University
2023
North Minzu University
2022-2023
Sun Yat-sen University
2011-2023
Localizing failure-inducing code is essential for software debugging. Manual fault localization can be quite tedious, error-prone, and time-consuming. Therefore, a huge body of research e orts have been dedicated to automated localization. Spectrum-based localization, the most intensively studied approach based on test execution information, may limited effectiveness, since element executed by failed tests not necessarily impact outcome cause failure. To bridge gap, mutation-based has...
Coverage-based fault localization has been extensively studied in the literature due to its effectiveness and lightweightness for real-world systems. However, existing techniques often utilize coverage an oversimplified way by abstracting detailed into numbers of tests or boolean vectors, thus limiting their practice. In this work, we present a novel coverage-based technique, GRACE, which fully utilizes information with graph-based representation learning. Our intuition is that can be...
In a virtual machine system, the scheduler within monitor (VMM) plays key role in determining overall fairness and performance characteristics of whole system. However, traditional VMM schedulers focus on sharing processor resources fairly among guest domains while leaving scheduling I/O missions as secondary concern. This would cause serious degradation make virtualization less desirable for I/O-intensive applications. order to eliminate bottleneck caused by delay, this paper proposes model...
Multimodal sentiment analysis aims to predict of language text with the help other modalities, such as vision and acoustic features. Previous studies focused on learning joint representation multiple ignoring some useful knowledge contained in modal. In this paper, we try incorporate sentimental words into fusion network guide multimodal Our method consists two components: shallow part aggregation part. For part, use crossmodal coattention mechanism obtain bidirectional context information...
In this paper, we present a new reinforcement learning (RL) algorithm called Distributional Soft Actor Critic (DSAC), which exploits the distributional information of accumulated rewards to achieve better performance. Seamlessly integrating SAC (which uses entropy encourage exploration) with principled view underlying objective, DSAC takes into consideration randomness in both action and rewards, beats state-of-the-art baselines several continuous control benchmarks. Moreover, propose...
Abstract Network embedding aims to map nodes in a network low-dimensional vector representations. Graph neural networks (GNNs) have received much attention and achieved state-of-the-art performance learning node representation. Using fundamental sociological theories (status theory balance theory) model signed networks, basing GNN on has become hot topic embedding. However, most GNNs fail use edge weight information models cannot be directly used weighted networks. We propose novel directed...
The China Conference on Knowledge Graph and Semantic Computing (CCKS) 2020 Evaluation Task 3 presented clinical named entity recognition event extraction for the Chinese electronic medical records. Two annotated data sets some other additional resources these two subtasks were provided participators. This evaluation competition attracted 354 teams 46 of them successfully submitted valid results. pre-trained language models are widely applied in this task. Data argumentation external also helpful.
In recent years, neural network models have been used in automated essay scoring task and achieved good performance. However, few studies investigated using the prompt information into network. We know that there is a close relevance between content topic. Therefore, topic can aid to represent relationship its score. That say, degree of high score will be higher while low less similar Inspired by this idea, we propose use similarity as auxiliary which concatenated final representation essay....
Current cross-prompt automated essay scoring (AES) is a challenging task due to the large discrepancies between different prompts, such as genres and expressions. The main goal of current AES systems learn enough shared features source target prompts grade well on prompt. However, because are captured based original prompt representation, they may be limited by being extracted directly essays. In fact, when representations two more similar, we can gain them. Based this motivation, in paper,...
We propose a packet-level model to investigate the impact of channel error on transmission control protocol (TCP) performance over IEEE-802.11-based multihop wireless networks. A Markov renewal approach is used analyze behavior TCP Reno and Impatient NewReno. Compared previous work, our main contributions are listed as follows: 1) modeling multiple lossy links, 2) investigating interactions among TCP, Internet Protocol (IP), media access (MAC) layers, specifically 802.11 MAC dynamic source...
We analyze the problem of buffer sizing for TCP flows in 802.11-based Wireless Mesh Networks. Our objective is to maintain high network utilization while providing low queueing delays. The complicated by time-varying capacity wireless channel as well random access mechanism 802.11 MAC protocol. While arbitrarily large buffers can utilization, this results Such delays may affect stability characteristics, and also increase other (including real-time flows) sharing buffer. In paper we propose...
The data collected by sensor networks often contain sensitive information and care must be taken to prevent that from being leaked malicious third parties, e.g., eavesdroppers. Under both the Neyman–Pearson Bayesian frameworks, we investigate strategy of defending against an informed greedy eavesdropper who has access all sensors' outputs via imperfect communication channels. Meanwhile, legitimate user, fusion center, is guaranteed achieve its desired detection performance. framework,...
For a connected graph G, the eccentric connectivity index (ECI) and first Zagreb eccentricity of G are defined as ( ) deg
Abstract With the rapid development of national economy, demand for electricity is also growing. Thermal power generation accounts highest proportion generation, and coal most commonly used combustion material. The massive has led to serious environmental pollution. It significant improve energy conversion efficiency reduce pollutant emissions effectively. In this paper, an extreme learning machine model based on improved Kalman particle swarm optimization (ELM-IKPSO) proposed establish...
Cognitive radio (CR) is a promising technique for improving the efficiency of utilizing precious spectrum. A cognitive network (CRN) testbed not only can verify concepts, algorithms, and protocols CR, but also reveal practical problems guide future research. Vision, architecture, intensive computing, frequency band shifting, security, ultra-wideband receiver potential applications are discussed in this paper. In support our arguments, experiments demonstrations reported as well.