Validating the use of off-the-shelf sensors for biometric data collection in affective computing

Abstract

Affective computing, or computing with mood and emotion, is likely to become an integral part of home robotics in the near future. Affective computing solutions have been developed across multiple sensor modalities, of which a popular, non-invasive solution is wearable biometric sensors. There are purpose-built devices for this task, however their price-point is largely prohibitive to their adoption in large, multi-user affective systems. This paper aims to address the issues with access to these devices by validating the use of off-the-shelf sensors for use in affective systems, with a mood prediction problem.