|
Descripción
|
|
|---|---|
| Thanks to great technological advances occurred in the latest years, particularly in ubiquitous computing, users are able to perform almost any kind of application in small mobile devices. The diversity and heterogeneity in culture, educational background, age, characterize these users. In many cases, online recommendation is not only about the services offered, but also related to actions to take in a certain situation. This would be an added value for mobile device users. However, contextaware recommendation is still an open issue. In this paper we focus on this problem, particularly on the challenge of providing dynamic recommendations, that are calculated locally in the device, using only the data produced by the user in each situation (location, time, action). We present a component based model in which the role of the controller component is underlined. The controller has to deal with the generation of models dynamically in a resource-aware environment. We also present a modular architecture in which the ubiquitous data mining components are integrated. | |
|
Internacional
|
Si |
|
Entidad
|
Workshop |
|
Lugar
|
|
|
Páginas
|
|
|
Referencia/URL
|
ecml/pkdd'08. UbiqKDD Workshop |
|
Tipo de publicación
|
actas del workshop |