Home > Research > Publications & Outputs > On-line trajectory classification

Electronic data

View graph of relations

On-line trajectory classification

Research output: Contribution in Book/Report/ProceedingsChapter

Publication date2003
Host publicationInternational Conference on Computational Science, Workshop on Scientific Visualisation and Human-Machine Interaction in a Problem-Solving Environment.
PublisherSpringer Verlag
Number of pages10
VolumeLNCS 2
<mark>Original language</mark>English


This study proposes a modular system for clustering on-line motion trajectories obtained while users navigate within a virtual environment. It presents a neural network simulation that gives a set of five clusters which help to differentiate users on the basis of efficient and inefficient navigational strategies. The accuracy of classification carried out with a self-organizing map algorithm was tested and improved to above 85% by using learning vector quantization. The benefits of this approach and the possibility of extending the methodology to the study of navigation in Human Computer Interaction are discussed.