A First Step Towards Nuance-Oriented Interfaces for Virtual Environments

Abstract

Designing usable interfaces for virtual environments (VEs) is not a trivial task. Much of the difficulty stems from the complexity and volume of the input data. Many VEs, in the creation of their interfaces, ignore much of the input data as a result of this. Using machine learning (ML), we introduce the notion of a nuance that can be used to increase the precision and power of a VE interface. An experiment verifying the existence of nuances using a neural network (NN) is discussed and a listing of guidelines to follow is given. We also review reasons why traditional ML techniques are difficult to apply to this problem.