Ensaio de Avaliação Preditiva para Utilizadores com Necessidades Especiais

Abstract

This paper addresses the prediction of likely user moves in a practical case study where user interaction is pre­dicted on the basis of lambda-parameters acquired from users with special needs playing a software game. A model was built to reproduce actions performed by such users in a 3D world to overcome the difficult set-up of tests due to user reduced availability, mobility and varying/erratic behaviour. User actions data was obtained from usability tests where users played a game requiring them to move around a virtual house and perform some ac­tivities. The tests were carried out with users with disabilities such as severe physical injuries, cerebral palsy and Down syndrome. A three level simulator was developed with data extracted from part o the data collected during experiments, the remaining raw experimental data serving as reference for validation. Data analysis cal­culated the λ-parameter required by the Poisson and Exponential Distributions used by the statistical approach adopted. The tests performed showed an excellent statistical flt of the simulated data to the reference data.