The ParetoSlider for Multi-criteria Decision Making by Non-Specialists
In Proceedings of the tenth Nordic conference on Human-computer interaction (NordiCHI'18). pages 1-10. 2018. à paraître.
Van Bao Nguyen, Yann Laurillau, Joëlle Coutaz, Gaëlle Calvary, Nadine Mandran, Raffaella Balzarini
Résumé
The ParetoSlider revisits the classical slider widget to help non domain-experts to find optimal compromises between tightly coupled, possibly conflicting, criteria. It is designed to make the tradeoff between criteria both observable and intelligible while hiding the complexity of an underlying optimization algorithm. As a case study, the ParetoSlider has been used in conjunction with an e-coach energy management system to interactively explore Pareto optimal solutions and choose the best compromise between comfort and financial cost. A preliminary semi-controlled empirical study with 16 participants using a 3-criteria ParetoSlider shows the suitability of this widget in terms of perceptual affordance, intelligibility, and exploration incentiveness.