Recommending Services in Pervasive Environments
Master thesis
Åpne
Permanent lenke
http://hdl.handle.net/11250/137496Utgivelsesdato
2010Metadata
Vis full innførselSamlinger
Sammendrag
The presence of mobile devices with network connectivity continues to increase in all parts of
the world, opening up direct channels to mobile users. The threat of information overload is
prevalent, and if not subjected to control in terms of legislation and deployment of more
intelligent and user-considerate systems it is likely to cause problems. This is particularly
important in the case of push-based advertisement systems where information must be of
relevance to the mobile user preferences, needs, and interest.
This paper proposes a solution and architecture of a consumer-oriented, push-based, context
aware location based advertisement system. The system is designed to operate in compliance
with the World Wireless Research Foundation (WWRF) guidelines for the future space of
information services: personalization, ambient awareness, and adaptability. The system is
intended to operate in pervasive environments.
The proposed system, called AdvertLBS, harvest knowledge about service providers and
consumers in terms of preferences, geographical and temporal locations. The system applies
knowledge about services, consumers, and learns context information in order to establish a
broad service recommendation context.
The AdvertLBS system adopts a hybrid approach to recommendation logic and proposes to
use a subset of suitable matching algorithms to produce service recommendations relevant to
the consumer’s interests and needs.
Beskrivelse
Masteroppgave i informasjons- og kommunikasjonsteknologi 2010 – Universitetet i Agder, Grimstad