Abstract

The wearable sensor market is currently one of the most rapidly growing area in consumer electronics. One of the most interesting application field for wearable sensors is early detection of diseases, seizures and attacks as it would save a huge amount of money from society. In order to train reliable recognition models for these purposes, a high quality training data needs to be collected. This study concentrates on challenges on gathering such data set based on our own experiences when collecting data using wearable sensors to detect migraine attacks.