Browsing by Author "Xiao Yan Shan"

UTSePress Research/Manakin Repository

Search UTSePress Research

Browse

My Account

Browsing by Author "Xiao Yan Shan"

Sort by: Order: Results:

  • K-Farthest-Neighbors-Based Concept Boundary Determination for Support Vector Data Description 
    Xiao Yan Shan; Liu Bo; Cao Longbing (ACM, 2010)
    Support vector data description (SVDD) is very useful for oneclass classification. However, it incurs high time complexity in handling large scale data. In this paper, we propose a novel and efficient method, named ...
  • A Learning Process Using SVMs for Multi-agents Decision Classification 
    Xiao Yan Shan; Deng Feiqi; Liu Bo; Liu Shouqiang; Luo Dan; Liang Guohua (IEEE CS digital library & Springer, 2008)
    The Third International Workshop on Agents and Data Mining Interaction (ADMIâ¿¿08) joint with the 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology. In addition, an EDITED ...
  • Multi-agent system for customer relationship management with SVMs tool 
    Xiao Yan Shan; Liu Bo; Luo Dan; Cao Longbing; Deng Feiqi; Hao Zhifeng (Inderscience, 2010)
    In this paper, we introduce multiple agents, knowledge discovery and data mining into customer relationship management (CRM) to set up the architecture of a multi-agent-based CRM system (MAB-CRM), and then use the SVMs-based ...
  • Multi-agent system for custom relationship management with SVMs tool 
    Xiao Yan Shan; Liu Bo; Luo Dan; Cao Longbing (Springer Berlin / Heidelb, 2008)
    Distributed data mining in the CRM is to learn available knowledge from the customer relationship so as to instruct the strategic behavior. In order to resolve the CRM in distributed data mining, this paper proposes the ...
  • Multi-sphere Support Vector Data for Outliers Detection on Multi-distribution Data 
    Xiao Yan Shan; Liu Bo; Cao Longbing; Wu Xindong; Zhang Chengqi; Hao Zhifeng; Cao Jie; Yang Fengzhao (IEEE Computer Society Press, 2009)
    SVDD has been proved a powerful tool for outlier detection. However, in detecting outliers on multi-distribution data, namely there are distinctive distributions in the data, it is very challenging for SVDD to generate a ...
  • One-class-based uncertain data stream learning 
    Liu Bo; Xiao Yan Shan; Cao Longbing; Yu Philip (SDM, 2011)
    This paper presents a novel approach to one-class-based uncertain data stream learning. Our proposed approach works in three steps. Firstly, we put forward a local kerneldensity-based method to generate a bound score for ...
  • Orientation Distance-based Discriminative Feature Extraction for Multi-Class Classification 
    Liu Bo; Xiao Yan Shan; Cao Longbing; Yu Philip (ACM, 2010)
    Feature extraction is an effective step in data mining and machine learning. While many feature extraction methods have been proposed for clustering, classification and regression, very limited work has been done on ...
  • A situation assessment approach using support vector machine as a learning tool 
    Lu Jie; Liu Bo; Zhang Guangquan; Hao Zhifeng; Xiao Yan Shan (InderScience Publisher, 2008)
    In order to assess a situation and support decision makers' awareness for the situation, this study first proposes a situation assessment model with mathematical description. It then develops a Support Vector Machine based ...
  • SMILE: A Similarity-Based Approach for Multiple Instance Learning 
    Xiao Yan Shan; Liu Bo; Cao Longbing; Yin Jie; Wu Xindong (IEEE Computer Society Conference Publishing Services (CPS), 2010)
    Multiple instance learning (MIL) is a generalization of supervised learning which attempts to learn useful information from bags of instances. In MIL, the true labels of the instances in positive bags are not always available ...
  • Vote-Based LELC for Positive and Unlabeled Textual Data Streams 
    Liu Bo; Xiao Yan Shan; Cao Longbing; Yu Philip (IEEE Computer Society Conference Publishing Services (CPS), 2010)
    In this paper, we extend LELC (PU Learning by Extracting Likely Positive and Negative Micro-Clusters) method to cope with positive and unlabeled data streams. Our developed approach, which is called vote-based LELC, works ...