Juggapong Natwichai

Publications (10)0 Total impact

• Article: Associative classification rules hiding for privacy preservation.

  Juggapong Natwichai, Xingzhi Sun, Xue Li
  IJIIDS. 01/2011; 5:246-270.
• Conference Proceeding: Data Reduction Approach for Sensitive Association Classification Rule Hiding.

  Juggapong Natwichai, Xingzhi Sun, Xue Li
  Database Technologies 2008. Proceedings of the Nineteenth Australasian Database Conference, ADC 2008, January 22-25, 2008, Wollongong, NSW, Australia; 01/2008
• Conference Proceeding: A Heuristic Data Reduction Approach for Associative Classification Rule Hiding.

  Juggapong Natwichai, Xingzhi Sun, Xue Li
  PRICAI 2008: Trends in Artificial Intelligence, 10th Pacific Rim International Conference on Artificial Intelligence, Hanoi, Vietnam, December 15-19, 2008. Proceedings; 01/2008
• Conference Proceeding: Data Quality in Privacy Preservation for Associative Classification.

  Nattapon Harnsamut, Juggapong Natwichai, Xingzhi Sun, Xue Li
  Advanced Data Mining and Applications, 4th International Conference, ADMA 2008, Chengdu, China, October 8-10, 2008. Proceedings; 01/2008
• Conference Proceeding: Hiding Sensitive Associative Classification Rule by Data Reduction.

  Juggapong Natwichai, Maria E. Orlowska, Xingzhi Sun
  Advanced Data Mining and Applications, Third International Conference, ADMA 2007, Harbin, China, August 6-8, 2007, Proceedings; 01/2007
• Source

  Available from: tu.ac.th

  Conference Proceeding: Efficient Semantically Equal Join on Strings.

  Juggapong Natwichai, Xingzhi Sun, Maria E. Orlowska
  Advances in Databases: Concepts, Systems and Applications, 12th International Conference on Database Systems for Advanced Applications, DASFAA 2007, Bangkok, Thailand, April 9-12, 2007, Proceedings; 01/2007
• Conference Proceeding: A reconstruction-based algorithm for classification rules hiding.

  Juggapong Natwichai, Xue Li, Maria E. Orlowska
  Database Technologies 2006, Proceedings of the 17th Australasian Database Conference, ADC 2006, Hobart, Tasmania, Australia, January 16-19 2006; 01/2006
• Article: Privacy-Preserving Data Mining of Classification Rules

  Juggapong Natwichai
  [show abstract] [hide abstract]
  ABSTRACT: Data sharing among collaborators is a common practice. On the one hand, shared data can be analysed by data mining techniques to provide useful patterns or knowledge for the collaborators. In the business domain, a new company may use shared data to discover the customer’s behaviour and thus improve their operations. While in the medical domain, epidemics can be determined more accurately when using the shared data from many hospitals. On the other hand, data sharing can pose a privacy threat for the data to be mined. With regard to data mining, there are two types of privacy of concern. Firstly, there could be sensitive or personal data which should not be disclosed. Secondly, some discoverable patterns being mined from the shared data could be considered as sensitive patterns. These sensitive patterns can be used to disclose personal data or give the other collaborators excessive information. These patterns should not be disclosed, or they should be “hidden” before data sharing takes place. Generally, to hide sensitive patterns in the data sharing scenarios, the data set to be shared needs to be modified until the interestingness of the sensitive patterns falls below the specific threshold. The modifications can cause damage to the statistical characteristics of the data set. This poses a challenging task, that is, how to preserve the characteristics of the modified data set required by the data recipient or “data usability” while the sensitive patterns are hidden? In this dissertation, our research focuses on the problem of sensitive pattern hiding, particularly for an important pattern type: classification rules. We propose algorithms to hide the sensitive rules while preserving the data usability; this can be categorised into two approaches: data reconstruction and data reduction. When shared data are to be used for the classification rule discovery and real data are not required, data reconstruction is a suitable approach. The approach introduces less side effects, that is, only non-sensitive rules are likely to be derived. We propose a data reconstruction framework to hide gain ratio-based classification rules. In this framework and to preserve the data usability, a set of remaining non-sensitive classification rules and gain-ratio information are used to reconstruct a new data set. For scenarios which real data are required, we address the problem through a data reduction approach, that is, by removing the selected records to hide the sensitive rules. Through this approach, the shared data are real data, although a subset of the original data. To preserve data usability, we investigate the impact on the data usability of the removals on a geometrical model. Observations of the impact are made in the model. By using the observations, the impact of a removal can be shown precisely without re-applying any classification algorithms on the data. We propose a number of algorithms by using the observations to hide sensitive associative classification rules in which they are categorised by data type - non-duplicate and duplicate data sets. Our proposed works are evaluated by standard real data sets with various characteristics. The evaluations are conducted both in terms of the effectiveness and the efficiency. For the effectiveness, sensitive rules with different features are randomly selected to be hidden. The hiding successfulness and the usability are evaluated. For the efficiency, our works are evaluated when sizes of the problem are scaled up, that is, higher interestingness of sensitive rules, higher number of sensitive rules, and larger size of data sets. From the evaluation results, our proposed works are not only effective - sensitive classification rules are hidden while data usability is well preserved - but also perform efficiently.
• Source

  Available from: psu.edu

  Article: Hiding Classification Rules for Data Sharing with Privacy Preservation

  Juggapong Natwichai, Xue Li, Maria E. Orlowska
  [show abstract] [hide abstract]
  ABSTRACT: In this paper, we propose a method of hiding sensitive classification rules from data mining algorithms for categorical datasets. Our approach is to reconstruct a dataset according to the classification rules that have been checked and agreed by the data owner for releasing to data sharing. Unlike the other heuristic modification approaches, firstly, our method classifies a given dataset. Subsequently, a set of classification rules is shown to the data owner to identify the sensitive rules that should be hidden. After that we build a new decision tree that is constituted only non-sensitive rules. Finally, a new dataset is reconstructed. Our experiments show that the sensitive rules can be hidden completely on the reconstructed datasets. While non-sensitive rules are still able to discovered without any side effect. Moreover, our method can also preserve high usability of reconstructed datasets.
• Source

  Available from: psu.edu

  Article: Knowledge maintenance on data streams with concept drifting

  Juggapong Natwichai, Xue Li
  [show abstract] [hide abstract]
  ABSTRACT: In this paper, we propose a method of hiding sensitive classification rules from data mining algorithms. Our idea is to reconstruct a dataset according to the classification rules that have been checked and agreed by the data owner for release to data sharing. Unlike other heuristic approaches, firstly, our method classifies a given dataset. Then, a set of classification rules are shown to the user. User then identifies the rules that are to be hidden. After that we generate a new decision tree that has only non-sensitive rules. A new dataset can then be reconstructed with no more and no fewer classification rules that can be derived. Our experiments show that this approach is efficient and effective.