Comfort conditions in buildings are no more a luxury than a necessity. Architects and designers are focussing their work on providing the best conditions of comfort in a building depending on the activity. The university gives students and teachers a space where they can share knowledge and experience. Besides, it should offer the best conditions for this process to work properly. That is why a lot of effort is put into bringing university occupants a proper space for working and learning. Despite all this effort, an evaluation of the comfort conditions in classrooms is necessary to ensure that indoor conditions can be adapted to occupant activity. One way to do this is to perform in situ measurements during teaching hours and compare the results to local standards. Besides the measures, we need occupants review regarding the Indoor Environment Quality (IEQ) trough survey. These data give us late feedback regarding the IEQ. therefore, with the different set of data acquired we intend to predict comfort conditions using different machine learning algorithms. A comparison between these algorithms will help us choose an adequate algorithm to estimate perceived comfort