ISABELA – A Socially-Aware Human-in-the-Loop Advisor System

Abstract

Currently, we are surrounded by all sorts of smart devices. Nevertheless, surprisingly, apart from e-health systems, few applications consider humans as active system players, as most applications/systems have the sole objective of providing some service to humans without considering their intents, actions, emotions, or social context. In this respect, they can be looked at as open loop, rather than integrating humans in a feedback loop, with potential to react to and change the environment. Systems that do consider human feedback are called Human-in-the-Loop Cyber-Physical-Systems (HiTLCPS), and these are now looked at as essential components of intelligent advisor systems (AS), i.e., systems that can take decisions and provide advice based the user environment and mood. In this paper, we present a HITLCPS AS system that combines online social networks (OSN) data with a variety of other data, such as environmental data and personal mobile devices data, in order to provide user advice. The paper details the system architecture, presents a data analysis from the collected dataset, provides results for some of the built models, and discusses future research directions.