User modeling in changeable environments

One of the challenges which need to be faced in the field of the information processing is the need to cope with huge amounts of data. There exist many different environments in which large quantities of information are produced. For example, in a command-line interface, a computer user types thousand of commands which can hide information about the behavior of her/his. However, processing this kind of streaming data on-line is a hard problem. This paper address the use of a dimensionality reduction perspective to classify streaming data and present a preliminary work for its adaptation to new data in changeable environments. This new extended proposal allows from a computational point of view the treatment of the new data without any penalty and evolving the previously analyzed data. Some results of the general technique are presented jointly with a comparison between the dimensionality reduction approach and using the full dataset. The results obtained show some of the advantages of using dimensionality reduction in streaming data and gives a new perspective to use it in changeable environments.