We developed mobile application -- called QoL Monitor -- to obtain health data from Health & Fitness Data Container (in our case, Google Fit was selected due to its slight integration complexity) and correlate this data with QoL questionnaires. The video on the side presents an overview of this tool.


This video presents our workflow to create a dataset correlating health measures and self-reported QoL questionnaires. Initially, the user must connect the native app of their wearable to sync data with the Google Fit platform. This connection is necessary to overcome the device heterogeneity issue. Once the data is registered in Google Fit, it is possible to extract it using a public API. Then, daily, QoL Monitor extracts the data recorded in Google Fit, anonymizes it by removing everything that can identify the user, cipher, and sends it to the cloud service. Finally, the app weekly requests the user to answer the QoL questionnaire to store it together with the health data.


Socio-demographic and anthropometric data are necessary to understand the characteristics of the users. The other raw data directly correlates with the health indicators proposed in this thesis. In addition, all of them can be obtained through common devices such as smart bands and smartwatches. Additionally, it is worth mentioning that the location data only stores the number of points that the user passed throughout the day, i.e., the application does not record the specific places. The same logic was applied to the WiFi Networks field. This strategy was adopted to preserve the users' privacy. In this video, we present details about the required permissions to collect data used to build our machine learning models.