In an increasingly complex world, with growing interactions between users and cyber-systems and an ever-increasing amount of information, machine learning techniques have become a fundamental tool for the EAC. Machine learning allows for more efficient use of data and facilitates the extraction of the hidden knowledge within it. The exploitation of data, in conjunction with data visualization, allows the EAC to speed up the insights discovery process, obtain more specific and valuable information, and incorporate the knowledge generated into the analysis of new input data to produce more dynamic visualization or predictive models.

Beyond pure data analytics, mass data processing techniques and automatic knowledge extraction have been an unprecedented force multiplier for technological progress in recent decades, making possible the development of new channels and mechanisms of communication between humans and computer systems. In recent years, the group has focused efforts on exploring new avenues of work at the intersection of machine learning with human-computer interaction to conceive new algorithmic solutions capable of recognizing and understanding both the activity and behavior of users (e.g., hand gesture recognition, body pose estimation, gaze tracking), as well as the different phenomena surrounding them to extract context-adapted knowledge (e.g., object detection, image recognition) and provide more efficient support for repetitive, tedious or highly complex processes.