Towards Self-explaining Intelligent Environments

Serge Autexier; Rolf Drechsler

In: IEEE 7th International Conference on Reliability, Infocom Technologies and Optimization (ICRITO'2018). IEEE Conference on Reliability, Infocom Technologies and Optimization (ICRITO-07), August 29-31, Noida, India, 2018.


Intelligent environments and assistants increasingly pervade and shape all areas of our daily life. They are characterised by dynamically varying numbers of coexisting, ideally cooperating but at least coordinated assistants and devices and users. This makes it difficult not only for technically inexperienced individual users to understand what is happening around them. It also does not help to know that the overall behaviour is correct in any sense whatsoever. From Human-ComputerInteraction research it is well known that systems providing explanations of their decision can improve the user’s trust in the systems. Illustrated by examples from our research on developing smart assistance in intelligent environments we motivate the need for self-explainability capabilities of intelligent environments. We shed a light on different aspects of self-explainability, motivate the need for explanations for the target user groups and provide criteria that are relevant to address for the design of selfexplaining intelligent environments. Index Terms—intelligent environments, assistance processes, human-computer-interaction, explanations


Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence