2014 | OriginalPaper | Chapter
Mining Incomplete Data with Attribute-Concept Values and “Do Not Care” Conditions
Authors : Patrick G. Clark, Jerzy W. Grzymala-Busse
Published in: Hybrid Artificial Intelligence Systems
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In this paper we present novel experimental results on comparing two interpretations of missing attribute values: attribute-concept values and “do not care” conditions. Experiments were conducted on 176 data sets, with preprocessing using three kinds of probabilistic approximations (lower, middle and upper) and the MLEM2 rule induction system. The performance was evaluated using the error rate computed by ten-fold cross validation. At 5% statistical significance level, in four cases attribute-concept values and in two cases “do not care” conditions performed better (out of 24 cases). At 10% statistical significance level, in five cases attribute-concept values and in three cases “do not care” conditions performed better. In the remaining cases the differences were not statistically significant.