AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
Our study finds that a close connection exists between attribute-oriented induction and rough set approach

Learning in Relational Databases: A Rough Set Approach

Computational Intelligence, no. 2 (1995): 323-338

Cited by: 640|Views31
EI

Abstract

Knowledge discovery in databases, or data mining, is an important direction in the development of data and knowledge-based systems. Because of the huge amount of data stored in large numbers of existing databases, and because the amount of data generated in electronic forms is growing rapidly, it is necessary to develop efficient methods ...More

Code:

Data:

Introduction
  • Knowledge discovery is the extraction of implicit, useful information from data. Knowledge discovery in databases is a form of machine learning which discovers interesting knowledge from databases and represents that knowledge in a high-level language.
  • The approach strives for efficiency in two aspects: knowledge-directed learning, and attribute-oriented induction.
  • The former is achieved by providing knowledge about the learning task, the concept hierarchies, and the expected rule forms.
  • The latter is achieved by attribute-oriented concept tree ascension.
  • These techniques substantially reduce the search space and improve the efficiency in a database learning process
Highlights
  • Knowledge discovery is the extraction of implicit, useful information from data
  • Knowledge discovery in databases is a form of machine learning which discovers interesting knowledge from databases and represents that knowledge in a high-level language
  • In our previous studies (Cai et al 1991; Han et al 1992), we developed an attribute-oriented induction approach for discovering knowledge in relational databases
  • Our study finds that a close connection exists between attribute-oriented induction and rough set approach
  • Our algorithms adopt the attributeoriented conceptual ascension technique combined with the rough set technique; attributeoriented induction provides a simple and efficient way for the generalization proccss, and rough set provides an effective tool to find out the reduction form of the generalized relation
Results
  • If the frequency ratio of a tuple is less than 5%, the tuple is cleared, otherwise the tuple is transformed to a logical rule.
Conclusion
  • The algorithms presented in this paper are based on the research in Cai et al (1991) and influenced by the ideas presented in Ziarko (1990).
  • The major difference of the approach from others is attribute-oriented vs tuple-oriented induction
  • The former techniques perform generalization, attribute by attribute, while the latter, tuple by tuple.
  • The authors' algorithms use rough set to analyze the attribute dependency and choose the reduct form of the best minimal subset for rule generalization.
  • 1991), which performs attribute-oriented generalization to extract knowledge rules in relational database, the authors are currently working on the implementation and testing of the DBDeci for the large database environment.
  • The authors plan to test it on a large database-that of the Research Grants Information System of Natural Science and Engineering Research Council of Canada-hope to report the results in the future
Funding
  • The authors are members of the Institute for Robotics and Intelligent Systems (IRIS) and wish to acknowledge the support of the Networks of Centres of Excellence Program of the Government of Canada, the Natural Sciences and Engineering Research Council, and the participation of PRECARN Associates Inc
Reference
  • DIEIwKicii, 'I. G., and R. S. MICHALSK1I9.83. A theory and methodology of inductive learning. In Machine learning: an artificial intelligence approach, vol. I. Edirrd by R. S. Michalski er a /.Morgan Kaufmann, San Mateo, CA, pp. 43-82.
    Google ScholarLocate open access versionFindings
  • GRZYMALA-BUSSJE.,W. 1988. Knowledge acquisition under uncertainty: a rough set approach. Journal of Intelligent and Robotic Systems, 1:3-16.
    Google ScholarLocate open access versionFindings
  • HAN,J., Y. CAI,and N. CERCONE. 1992. Knowledge discovery in databases: an attribute-oriented approach. In Proceedings of the 18th VLDB Conference, Vancouver. B.C., Canada, pp. 340-355.
    Google ScholarLocate open access versionFindings
  • MANAGOM,. V., and Y. KODRATOFF1.987. Noise and knowledge acquisition. In Proceedings of the 10th International Joint Conference on Artificial Intelligence, Milan, Italy, pp. 348-354.
    Google ScholarLocate open access versionFindings
  • MICKALSKRI., S. 1983. A theory and methodology of inductive learning. In Machine learning: an artificial intelligence approach, vol.
    Google ScholarLocate open access versionFindings
  • 1. Edited by R. S. Michalski er al. Morgan Kaufmann, San Mateo, CA, pp. 4182. PAWLAKZ,. 1982. Rough sets. International Journal of Information and Computer Science, 11(5):341-356.
    Google ScholarLocate open access versionFindings
  • PAWLAKZ,. 1990. Anathomy of conflicts. In ICS Research Report 11/92. Warsaw University of Technology, Warsaw, Poland.
    Google ScholarFindings
  • ZIARKOW,. 1990. The discovery, analysis, and representation of data dependencies in databases. In Knowledge discovery in databases. Edited bv G. Piatetsky-Shapiro and W. J. Frawley. AAAIMIT Press, Cambridge, MA, pp. 2 13-228.
    Google ScholarLocate open access versionFindings
  • 990. Interactive mining of irregularities in databases. I n Knowledge discovery in databases. Editedbv G. Piatetsky-Shapiro and W. J. Frawley. AAAUMIT Press, Cambridge, MA, pp. 31-54.
    Google ScholarFindings
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科