Table 3. Comparison of precision for migraine classifications reported in previous studies.

Reference study	Classification model	Precision
Migraine diagnosis support system based on classifier ensemble ( Jackowski et al., 2014)	LAD Tree	75.9%
Automatic diagnosis of primary headaches by machine learning methods ( Krawczyk et al., 2013)	Random Forest	81%
Analysis of repetitive flash stimulation frequencies and record periods to detect migraine using artificial neural network ( Akben et al., 2012)	ANN	83.3%
Classification of multi-channel EEG signals for migraine detection ( Akben et al., 2016)	SVM	85%
Effect of photic stimulation for migraine detection using random forest and discrete wavelet transform ( Subasi et al., 2019)	Random Forest	85.95%
Analysis of Artificial Neural Networks Models, for a System of Diagnoses of Migraines with Aura and without Aura ( De la Hoz & Rúa et al., 2014)	ANN	91.04%
A clinical decision support system for the diagnosis of probable migraine and probable tension-type headache based on case-based reasoning ( Yin et al., 2015)	CBR	93.14%
This study	ANN (complete)	97%
This study	ANN (reduced)	98%

LAD - least absolute deviations, ANN – artificial neural network, SVM – support vector machine, CBR – case-based reasoning