Title :
Query ranking strategies in probabilistic expert systems
Author :
Yuzuguzel, Hidir ; Cemgil, A.T. ; Anarim, Emin
Abstract :
The number of features are quite high in many fields. For instance, the number of symptoms are around thousands in probabilistic medical expert systems. Since it is not practical to query all the symptoms to reach the diagnosis, query choice becomes important. In this work, 3 query ranking strategies in probabilistic expert systems are proposed and their performances on synthetic data are evaluated.
Keywords :
medical diagnostic computing; medical expert systems; probability; query processing; medical diagnosis; probabilistic expert systems; probabilistic medical expert systems; query ranking strategies; Conferences; Entropy; Expert systems; Inference algorithms; Probabilistic logic; Sequential diagnosis; Signal processing; medical diagnosis; relative-entropy; sequential diagnosis;
Conference_Titel :
Signal Processing and Communications Applications Conference (SIU), 2014 22nd
Conference_Location :
Trabzon
DOI :
10.1109/SIU.2014.6830450
