HMI Evaluation Framework
This repository is devoted to contain the results of the research in the field of quantitative evaluation for the design of advanced HMI systems.
The DARWINIST project
The research is framed into the aDversarial scenArios geneRation With dIgital twiNs In induSTry (DARWINIST) project, whose aim is to explore novel architectural and methodological methods for the design of Digital Twins with a special focus on the integration between data-driven and model-based approaches. The project is funded by Università della Campania "Luigi Vanvitelli".
The ANDROIDS project
The research is also supported by AutoNomous DiscoveRy Of depressIve Disorder Signs (ANDROIDS)project, which is a past closed funded by Università della Campania "Luigi Vanvitelli" by means of the V:alere funding programme. The project investigates the core features of human interactions to model cognitive and emotional processes. The aim is to design and implement autonomous systems and algorithms to detect early signs of mood changes and depressions through the analyses of interaction exchanges. ANDROIDS aims to combine diverse sources of information as well as to understand psychological dynamics from behavioural data.
More details at: https://www.psicologia.unicampania.it/android-project
Long-term objective
In the context of the ANDROIDS project, this research has the long term goal to define and implement a framework for the design and the quantitative evaluation of HMIs in next generation smart and emotion-aware virtual agents. To this aim, the main steps of the roadmap to accomplishing the research are the definitions of:
- an emotion-aware digital twin of the user;
- a data-driven model of the interaction between the human and the virtual agent;
- mechanisms and guidelines for the manual editing of the user model;
- mechanisms and automations for the quantitative analysis of composed user-interaction model.
Current status
Up to this date, the points 1 and 2 of the previous list has been investigated and first results of the research are reported in this repository.
Content
The repository is composed of the following sub-folders:
- MIDas4CS23_replication: containing the material for the paper referenced in [1]
People
- Stefano Marrone - associate professor in Computer Science, Dipartimento di Matematica e Fisica, Università della Campania "Luigi Vanvitelli", stefano.marrone@unicampania.it
- Laura Verde - assistant professor in Computer Science, Dipartimento di Matematica e Fisica, Università della Campania "Luigi Vanvitelli", laura.verde@unicampania.it
References
- [1] Lelio Campanile, Roberta De Fazio, Michele Di Giovanni, Stefano Marrone, Fiammetta Marulli, Laura Verde; Inferring Emotional Models from Human-Machine Speech Interactions; submitted to First Workshop on the Modelling and Implementation of Digital Twins for Complex Systems (MIDas4CS 2023).
- [2] De Biase, M.S., Marrone, S., Marulli, F.; Automatic Generation of Smart Human-Machine Interfaces; Proceedings of the 2020 IEEE International Conference on Human-Machine Systems, ICHMS 2020, art. no. 9209522; DOI: 10.1109/ICHMS49158.2020.9209522