National genetic evaluation results form the basis of Interbull services. The current method for quality assurance is mainly determined by the consistency of consecutive evaluation results and is based on thorough statistical examination (Klei et al., 2002). In a separate project, national genetic evaluation programs are being tested on simulated data sets with known properties (Taubert et al., 2002). Datamining (DM) offers an alternative way to examine data and extract valuable information (Han and Kamber, 2000), potentially leading to inference on data quality. In a recent progress report, the development of a DM algorithm for the analysis of national genetic evaluation results was presented (Banos et al., 2003). Data quality was assessed by subjective inspection of DM results. The present study introduces a method to evaluate DM application results with objective criteria leading, when necessary, to the automatic issuing of warnings or alarm signals.
[1]
F Fikse,et al.
A method for verifying genetic evaluation results
,
2002
.
[2]
Jiawei Han,et al.
Data Mining: Concepts and Techniques
,
2000
.
[3]
Periklis Mitkas,et al.
Quality control of national genetic evaluation results using data-mining techniques; a progress report
,
2003
.
[4]
J. Ross Quinlan,et al.
C4.5: Programs for Machine Learning
,
1992
.
[5]
Henner Simianer,et al.
The Interbull audit project part II: Development of a program for auditing breeding value estimation programs
,
2002
.
[6]
M. Kendall.
Elementary Statistics
,
1945,
Nature.