An attribute-based classification by threshold to enhance the data matching process

Main Article Content

Maria Del Pilar Angeles
Carlos G. Ortiz Monreal

Abstract

The problem of detection and classification of extensional inconsistencies during data integration of disparate data sources affects business competitiveness. A number of classification methods have been utilized until now, but there still some work to do in terms of effectiveness and performance. The paper shows the proposal, implementation, and evaluation of a new classification algorithm called Attribute-based Classification by Threshold that overcomes the disadvantages of the Threshold-based Classification. We have carried aout an evaluation of quality of the data matching process by comparing Threshold-based Classification, Farthest First and K-means against the proposed algorithm. The Attribute-based Classification by Threshold has a better performance than the rest of the unsupervised classification methods.

 

PLUM ANALYTICS

Article Details

How to Cite
Angeles, M., & Monreal, C. G. (2019). An attribute-based classification by threshold to enhance the data matching process. Journal of Applied Research and Technology, 17(4). https://doi.org/10.22201/icat.16656423.2019.17.4.861
Author Biographies

Maria Del Pilar Angeles

Instituto de Investigaciones Aplicadas y en Sistemas, Universidad Nacional Autonoma de Mexico

Carlos G. Ortiz Monreal

Oracle Mexico