Automatic Classification for the Identification of Relationships in a Meta-Data Repository
For a large company a prototype for automatic detection of similar objects in database systems has been developed. This task has been accomplished by transferring the database object classification problem into a text classification problem and applying standard classification algorithms. Although the data provided for the task did not look promising due to the small number of positive examples, the results turned out to be very good.
KeywordsVector Representation Object Type Data Repository Relationship Type Database Object
Unable to display preview. Download preview PDF.
- 1.Beuster, G.: MIC — A System for Classification of Structured and Unstructured Texts. Master’s thesis, University Koblenz (2001), http://www.gb/papers/thesis_mic/mic.pdf
- 2.Bouguettaya, A., Benatallah, B., Elmagarmid, A.K.: Interconnecting Heterogeneous Information Systems. Kluwer Academic Publishers, Dordrecht (1998)Google Scholar
- 4.Marco, D.: Building and Managing the Meta Data Repository: A Full Lifecycle Guide. John Wiley & Sons, Chichester (2000)Google Scholar
- 6.Mitchell, T.M.: Machine Learning. McGraw-Hill International Editions (1997)Google Scholar
- 7.Quinlan, J.: Discovering rules by induction from a large collection of examples. In: Michie, D. (ed.) Expert systems in the Micro-Electronic Age, pp. 168–201. Edinburgh University Press, Edinburgh (1979)Google Scholar
- 8.Rumelhart, D.D., Hinton, G.E., Williams, R.J.: Learning representations by backpropagating errors. Nature, 533–536 (1986)Google Scholar