engleski

Comparison of two inductive algorithms in diagnosing thyroid functional status

The aim of this paper is to compare results of two algorithms, Assistant-algorithm and ILLM (Inductive Learning by Logic Minimisation), applied on thyroid function laboratory diagnostics data. Data base comprised results of the routine assays performed at Clinical Hospital "Sestre Milosrdnice", Zagreb. Overall accuracy of Assistant-algorithm and ILLM-algorithm predictions were 91.7 % and 90.6% respectively. Considering overall accuracies and all specific accuracies in both algorithms it can be seen that ILLM algorithm gives slightly better results, although both algorithms gave medically sound decision tructure.

machine learning; thyroid functional status

nije evidentirano