A5101485888- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Simulation and Modeling Applications
- Data Management and Algorithms
- Cloud Computing and Resource Management
- Peer-to-Peer Network Technologies
- Fault Detection and Control Systems
- Complex Systems and Time Series Analysis
- Advanced Text Analysis Techniques
- Mobile Ad Hoc Networks
- Vehicular Ad Hoc Networks (VANETs)
- Robotic Path Planning Algorithms
- Opportunistic and Delay-Tolerant Networks
- Graph Theory and Algorithms
- Optimization and Search Problems
- Advanced Image Processing Techniques
- Advanced Data Storage Technologies
- Complex Network Analysis Techniques
- Scheduling and Optimization Algorithms
- Distributed and Parallel Computing Systems
- Advanced Vision and Imaging
- Image Processing Techniques and Applications
- Video Surveillance and Tracking Methods
- Network Security and Intrusion Detection
- Advanced Database Systems and Queries
PLA Army Engineering University
2011-2023
Kunming University of Science and Technology
2005-2022
Luoyang Institute of Science and Technology
2019
National University of Defense Technology
2012-2015
Xiangyang Central Hospital
2014-2015
Jiangxi Normal University
2013-2014
Yancheng Teachers University
2003
The satellite fault diagnosis has an important role in enhancing the safety, reliability, and availability of system. However, problem enormous parameters multiple faults makes a challenge to diagnosis. interactions between misclassifications from will increase false alarm rate negative rate. On other hand, for each fault, there is not enough data training. To most classification algorithms, it degrade performance model. In this paper, we proposed improving SVM based on hybrid voting...
The similarity search is one of the fundamental components in time series data mining, e.g. clustering, classification, association rules mining. Many methods have been proposed to measure between series, including Euclidean distance, Manhattan and dynamic warping (DTW). In contrast, DTW has suggested allow more robust be able find optimal alignment series. However, due its quadratic space complexity, not suitable for large datasets. improving algorithms databases, such as approximate or...
BACKGROUND: The widespread access to portable medical devices or new personal is boosting the amount of biomedical data.These provide a growing massive data that far exceeds analytical ability professional doctor.The computer-assisted analysis has become an essential tool in medicine diagnosis.OBJECTIVE: Due advantages discrete, noise elimination and dimensionality reduction, symbolic representation attracted great interest.The symbolization results efficiently performing at mining, such as...
Column Family has been used in NoSQL Database, and we have to think over how divide the designing of Database. But dividing column family is similar vertical partitioning distributed database. And there are some object functions database system, but they do not fit In this paper will propose an objective function for class The derived generalizes subsumes earlier work on scenarios Furthermore, approach proposed shown be useful evaluate cost family.
In recent years, MapReduce parallel computing model has gained lots of attentions from industry and academia. Google, Yahoo, Facebook, etc., it played a very good effect, which greatly simplifies the design large-scale data-intensive applications. MapReduce-based systems were originally used to manage massive unstructured semi-structured data, for example: generate inverted index, calculate web page rank, log analysis, etc. Therefore, current don't give more considerations optimization...
Scheduling algorithms place a crucial role in MapReduce systems. Several recent scheduling algorithms, however, are all under Job-Task model which makes task confined, leading to poor preference such as data locality, scan sharing and etc. These characteristics very important heuristics on intensive computing helpful improving system throughput. In this paper, we firstly design novel termed Tasks-Job overcome the above issues. Furthermore, propose locality aware algorithm improve...
With the emerging of large database and explosion mass data, people need to extract useful information, knowledge from huge database, improve utilization rate information further. Partitioning technologies allows users divisive a big table into some smaller can manage partitions easier, so it solve problems data. This paper discussed Partial-MAX/MIN query optimization for partitioned tables. We introduce Rank Bisection Partition Tree (RBP-T) structure improving efficiency this class query....
The heart sound signal is a reflection of and vascular system motion. Long-term continuous electrocardiogram (ECG) contains important information which can be helpful to prevent failure. A single piece long-term ECG recording usually
One fundamental challenge for Alert Correlation(AC) is to learn attack strategies. Attack graph one of the most commonly used models describe patterns(strategies), however, generating technology far from practical. There are two general approaches generate graphs. The first uses based search find paths possible attacks, such as model checking, assumes that premises and consequences attacks known. And other makes use statistical method, frequent sequence mining, try relationship in dimension...
Background: Wound infection (WI) is a disease in which pathogenic bacteria invade and multiply wound after trauma or surgery, causing systemic inflammatory response. WI triggers an immune response the body, resulting inflammation tissue damage, as well slowing down healing process. The traditional Chinese medicine prescription of Wuwei Xiaodu Drink (WWXDD) has been widely used clinical practice with good results. However, there no high-level evidence to support this result. purpose study was...
An improved A* search algorithm was proposed to solve Path-optimization problems in emergency actions. By altering the direction and reforming evaluate function, make it more suitable for dealing with large problems. Through instance validation of command air-defense system a certain city, get better results.
A novel abnormity detection method is presented which combines the low‐rank structured sparse representation and reduced dictionary learning. The multi‐scale three‐dimensional gradient used as low‐level feature by encoding spatiotemporal information. group of dictionaries learnt approximation based on sparsity property video sequence. contribution this study three‐fold: (i) normal clusters can be represented effectively are nature data; (ii) size determined adaptively learning according to...
Aims to the routing protocol of Bus Vehicle Network with characteristics cyclical movement, we put forward Signal-Single & Multi hop Routing (SSMHR) protocol, which can solve problem signal attenuation on urban environment and uncertain next forwarding multi-hop. Link Survival Mode considering will be established in Single-hop select encounter nodes as candidate set. Discrete Stochastic Dynamic Programming Delay Probability Model Multi-hop using historical data bus movement. The two...