Development of a Mobile Mood Detector based on Machine Learning Techniques

Jorge García-Castaño. (2020). Development of a Mobile Mood Detector based on Machine Learning Techniques. Trabajo Fin de Titulación (TFM). Universidad Politécnica de Madrid, ETSI Telecomunicación.

Abstract:
According to data published by the World Health Organization (WHO), more than two million people suffer from mental disorders (such as depression) in Spain, which is 5.2% of the population of our country. These data have increased considerably in recent years and, from 2005 to 2015, the number of people suffering from depression has increased by 18.4%. Showing numbers at global level, more than 320 million people suffered from depression in 2015. In the case of anxiety, 264 million people suffered from it globally in the same year, which is 3.6% of the world’s population. To conclude with WHO statistics, about 788,000 people commit suicide each year because of depression or other mental disorder. In addition, numerous scientific studies confirm that engaging in daily sports activities has a positive influence on people’s health and mood. In other words, daily physical activity is healthy because it cures and prevents illness. However, this relationship can also be understood as a subjective interpretation of health, which is nothing more than personal well-being. In this project, an Android application is developed, which constantly monitors the user’s mood. In addition, the application collects data related to instant messaging con- versations (such as Whatsapp), as well as data related to the user’s daily physical activity. Based on the data obtained, machine learning algorithms will be applied in order to predict and classify the user’s mood.