A5101999870- Bayesian Modeling and Causal Inference
- Anomaly Detection Techniques and Applications
- Data Quality and Management
- Time Series Analysis and Forecasting
- Mobile Agent-Based Network Management
- Machine Learning and Data Classification
- Network Security and Intrusion Detection
- Software System Performance and Reliability
- Web Data Mining and Analysis
- Artificial Intelligence in Healthcare
- Education and Work Dynamics
- Data Mining Algorithms and Applications
- Cloud Computing and Resource Management
- Advanced Graph Neural Networks
- Higher Education and Teaching Methods
- Power Systems and Technologies
- Complex Network Analysis Techniques
Changzhou Institute of Technology
2009-2019
Purpose With the introduction of graph structure learning into service classification, more accurate structures can significantly improve precision classification. However, existing methods tend to rely on a single information source when attempting eliminate noise in original and lack consideration for generation mechanism. To address this problem, paper aims propose estimation neural network-based classification (GSESC) model. Design/methodology/approach First, method uses local smoothing...
Bayesian network classifiers (BNCs) have demonstrated competitive classification performance in a variety of real-world applications. A highly scalable BNC with high expressivity is extremely desirable. This paper proposes Redundant Dependence Elimination (RDE) for improving the and k-dependence classifier (KDB). To demonstrate unique characteristics each case, RDE identifies redundant conditional dependencies then substitute/remove them. The learned personalized Classifier (PKDB) can...
Statistical techniques play a crucial role in anomaly detection. Although they usually are simple and can be trained unsupervised, face three challenges: parametric rely on the assumption that data meet special distribution; existing Histogram-based only take account of individual attribute, which cannot capture interactions between different attributes; some statistical still need labeled for training or validation. In order to overcome these drawbacks, this paper proposes statistic method...
Machine learning techniques have shown superior predictive power, among which Bayesian network classifiers (BNCs) remained of great interest due to its capacity demonstrate complex dependence relationships. Most traditional BNCs tend build only one model fit training instances by analyzing independence between attributes using conditional mutual information. However, for different class labels, the relationships may be rather than invariant when take values, result in classification bias. To...
The superparent one-dependence estimators (SPODEs) is a popular family of semi-naive Bayesian network classifiers, and the averaged (AODE) provides efficient single pass learning with competitive classification accuracy. All SPODEs in AODE are treated equally have same weight. Researchers proposed to apply information-theoretic metrics, such as mutual information or conditional log likelihood, for assigning discriminative weights. However, while dealing different instances independence...