A5101595779- Spam and Phishing Detection
- Cloud Computing and Resource Management
- Complex Network Analysis Techniques
- Network Security and Intrusion Detection
- Access Control and Trust
- Data Quality and Management
- Privacy-Preserving Technologies in Data
- Hydraulic flow and structures
- Privacy, Security, and Data Protection
- Internet Traffic Analysis and Secure E-voting
- Underwater Vehicles and Communication Systems
- Big Data and Business Intelligence
- Anomaly Detection Techniques and Applications
- Simulation and Modeling Applications
- Energy Efficient Wireless Sensor Networks
- Digital Media and Visual Art
- Robotic Locomotion and Control
- Energy Harvesting in Wireless Networks
- Data Analysis with R
- Sentiment Analysis and Opinion Mining
- User Authentication and Security Systems
- Control and Dynamics of Mobile Robots
- Peer-to-Peer Network Technologies
- Text and Document Classification Technologies
- Belt Conveyor Systems Engineering
Chinese Academy of Sciences
2018-2024
Institute of Modern Physics
2018-2024
Jiangnan University
2024
Shandong Institute of Automation
2024
Shandong University
2018-2023
Illinois State University
2017-2023
Chongqing University
2009-2022
Huazhong University of Science and Technology
2008-2022
Weihai Science and Technology Bureau
2018-2020
Guangxi University
2019
Sentiment analysis or opinion mining is one of the major tasks NLP (Natural Language Processing). has gain much attention in recent years. In this paper, we aim to tackle problem sentiment polarity categorization, which fundamental problems analysis. A general process for categorization proposed with detailed descriptions. Data used study are online product reviews collected from Amazon.com. Experiments both sentence-level and review-level performed promising outcomes. At last, also give...
The Jiangmen Underground Neutrino Observatory (JUNO) is proposed to determine the neutrino mass hierarchy using an underground liquid scintillator detector. It located 53 km away from both Yangjiang and Taishan Nuclear Power Plants in Guangdong, China. experimental hall, spanning more than 50 meters, under a granite mountain of over 700 m overburden. Within six years running, detection reactor antineutrinos can resolve at confidence level 3-4$\sigma$, oscillation parameters...
Abstract Sentiment analysis is recognized as one of the most important sub-areas in Natural Language Processing (NLP) research, where understanding implicit or explicit sentiments expressed social media contents valuable to customers, business owners, and other stakeholders. Researchers have that generic extracted from textual are inadequate, thus, Aspect Based Analysis (ABSA) was coined capture aspect toward specific review aspects . Existing ABSA methods not only treat analytical problem...
Like how useful weather forecasting is, the capability of or predicting cyber threats can never be overestimated. Previous investigations show that attack data exhibits interesting phenomena, such as long-range dependence and high nonlinearity, which impose a particular challenge on modeling rates. Deviating from statistical approach is utilized in literature, this paper we develop deep learning framework by utilizing bi-directional recurrent neural networks with long short-term memory,...
Routing plays a critical role in data transmission for underwater acoustic sensor networks (UWSNs) the internet of things (IoUT). Traditional routing methods suffer from high end-toend delay, limited bandwidth, and energy consumption. With development artificial intelligence machine learning algorithms, many researchers apply these new to improve quality routing. In this paper, we propose Q-learning-based multi-hop cooperative protocol (QMCR) UWSNs. Our can automatically choose nodes with...
The mechanisms of locomotion in mammals have been extensively studied and inspire the related researches on designing control architectures for legged robots. Reinforcement learning (RL) is a promising approach allowing robots to automatically learn policies. However, careful reward-function adjustments are often required via trial-and-error until achieving desired behavior, as RL policy behaviors sensitive rewards. In this paper, we draw inspiration from rhythmic animals propose new...
Online banking authentication plays an important role in the field of online security. In past years, a number methods, including password token, short message password, and USB have been developed for authentication. this paper, we introduce new protocol banking. Our approach enhances performance robustness against various attacks by using mobile phones to store digital certificate clients. We provide attack analysis illustrate strength protocol.
Social computing can be broadly defined as computational facilitation of social studies and human dynamics well design use information communication technologies that consider context.Novel services applications developed based on have profoundly affected nowadays people's life.However, an emerging research field, it gathers numerous researchers who tend to contribute cutting-edge work computing.In this paper, we provide a brief overview towards the state art computing.
For a given source-destination pair in multi-hop underwater acoustic sensor networks (UASNs), an optimal route is the one with lowest energy consumptions that usually consists of same relay nodes even under different transmission tasks. However, this will lead to unbalanced payload UASNs and accelerate loss working ability for entire system. In paper, we propose node balanced ant colony cooperative routing (PB-ACR) protocol UASNs, through combining algorithm transmission. The proposed PB-ACR...
Quadruped locomotion is a challenging task for learning-based algorithms. It requires tedious manual tuning and difficult to deploy in reality due the gap. In this paper, we propose quadruped robot learning system agile which does not require any pre-training works well various real-world terrains. We introduce hierarchical framework that uses reinforcement as high-level policy adjust low-level trajectory generator better adaptability terrain. compact observation action space of it on host...
In the early 2000s, Texas Department of Transportation funded several research projects to examine unit hydrograph and rainfall hyetograph techniques for hydrologic design in estimation flows stormwater drainage systems. A consortium comprised Lamar University, Tech University Houston, U.S. Geological Survey (USGS), was chosen techniques. Rainfall runoff data collected by USGS at 91 streamflow-gaging stations formed a basis research. These were as part small-watershed urban watershed studies...
The objective of this study was to develop a machine learning (ML) application determine the optimal dosage sodium hypochlorite (NaOCl) curtail corrosion and odor by H2 S in headworks water resource recovery facility (WRRF) without overly consuming volatile fatty acids (VFAs) that are essential for enhanced biological phosphorus removal. Given highly diverse datasets available, three subproblems were formulated, cascaded ML modules developed accordingly. final models, chosen based on...
Understanding sentiments embeded in online consumer reviews can assist a variety of decision making processes. Recent studies have extended sentiment analysis to fine-grained level, such as toward different features or characteristics products/services, which is known Aspect Based Sentiment Analysis (ABSA). In this study, we propose transfer learning based multi-label classification approach, building on pre-trained advanced deep model. We evaluate the proposed approach an experiment by...
Convolutional neural networks are widely adopted for solving problems in image classification. In this work, we aim to gain a better understanding of deep learning through exploring the miss-classified cases facial and emotion recognitions. Particularly, propose backtracking algorithm order track down activated pixels among last layer feature maps. We then able visualize features that lead miss-classifications, by applying tracking algorithm. A comparative analysis reveals recognition,...
Despite the significant research achievements on study of communities, how to maintain a benign social environment for community, as problem, has not received much attention. Current existing malicious activity detecting mechanisms are subject limitation underlying online environment. However, we found that information plays an important role in terms socialisation. Malicious activities, like diffusing unreliable and providing inappropriate critics, can severely impact situation community’s...
Modeling cyber risks has been an important but challenging task in the domain of security, which is mainly caused by high dimensionality and heavy tails risk patterns. Those obstacles have hindered development statistical modeling multivariate risks. In this work, we propose a novel approach for relies on deep learning extreme value theory. The proposed model not only enjoys accurate point predictions via also can provide satisfactory quantile Both simulation empirical studies show that very...
Social computing is the backbone of increasing socialized web applications, which more and people are tending to intensively rely on. Trust a critical element for social that has been involved into many computational systems. It topic intriguing numerous studies. In this paper, we present novel Computational Framework based on perspective information sharing. Our able quantify trust value given message. To our best knowledge, first time measured under reliability Entropy. We also explain...