The challenge of determining an appropriated indoor temperature in office buildings for most occupants is a difficult task due to the large number of parameters influencing the thermal preference of building occupants. Current Machine Learning techniques assist in determining and characterising the optimal values of the environmental parameters influencing the indoor thermal comfort of building occupants, helping to optimise the temperature setpoint of the heating, ventilation and air conditioning systems, adapting it to the real preferences of the occupants, which translates into direct savings in the energy consumed by the occupants. Therefore, this study proposes the implementation of Artificial Neural Networks to obtain comfort zones for the occupants of a mixed-mode office building located in Mediterranean climate.