This emotion detection system uses a Raspberry Pi and two non-invasive sensors- Body Temperature and Electrocardiogram sensor- to collect data from a user and send it to the IoT analytics platform “Thingspeak”. The initially generated data is used to train a machine learning model that uses the Random Forest Classifier predominantly and returns the emotion value of the user depending on his/her collected body temperature and ECG value. This result is displayed on an interactive dashboard.