Location-Aware Information Services using User Profile Matching Alisa Devli´ c University of Zagreb, FER Unska 3, Zagreb, Croatia (HR-10000) Email: alisa.devlic@fer.hr Gordan Jeˇ zi´ c University of Zagreb, FER Unska 3, Zagreb, Croatia (HR-10000) Email: gordan.jezic@fer.hr Abstract— User profiles are used to integrate contextual information about mobile users and objects/devices in their environment. They represent structured knowledge used by a system to provide relevant information set to the user. Location-aware information services have to adapt their functionality to a constantly changing user location and preferences. The adaptation is based on a content ontology, that contains concepts for describing user personal information, device characteristics, presence information and location-based subscription preferences. User profiles are generated for every user’s applied terminal and matched to enable adaptation and dissemination of location-aware content. The proposed architecture is sketched and the initial implementation prototype of the system is described in the article. I. I NTRODUCTION The idea of context-aware computing came from the growing number of wireless and mobile devices used in the user everyday life and the user need to be served with timely delivered information anyplace, anytime and on any device, while on the move. Users want to access the same content and services on various devices, e.g. laptops, handheld devices, mo- bile phones, that have different processing capabilities, memory capacity, screen size and support different media formats. Users may also wish to change the terminal during the session and the service task is to recognize the user and enable him a smooth transition to the other device and the continuity of the service use. The mobile context of the user should be tracked and the awareness of own context is important for applications to be able to adapt their behavior according to the present situation and the conditions in the environment. This way a new set of information services, called context-aware information services, will become possible on mobile devices. User location is one of the main drivers for the context- aware information. Location-aware information services are services that provide the user with the information set that is related to his current position. Because user position can change between the requests for information, different information sets can be generated for different user positions. There are many positioning methods that can locate the user terminal, such as GPS, A-GPS, Cell ID+TA, E-OTD. The problem that arises is how to ag- gregate the various forms of location information that these methods produce and give it the user-understandable meaning. In this article we define the term landmark to denote a located user position with the given location name and propose a way to integrate different forms of location information to be used in location-aware information services. In addition we propose a way to integrate the knowl- edge about the user context and describe the contextual information in the user profile form. The proposed profile consists of attributes concerning characteristics of the user’s currently used terminal, his location and presence information, and subscription preferences. We assume that the user may prefer to receive the desired content either on the current location or some other location of interest (defined landmark). For that reason we define two kinds of location-based subscriptions: landmark-based and current location-based subscription. The information from the user profile is used to adapt and disseminate the content to the user currently used terminal. For the implementation purposes we have devel- oped an initial prototype built on the proposed architecture consisting of the client application that interacts with dis- tributed server components through the web application, having the following functionalities: profile administra- tion, location management, and content adaptation and dissemination. II. CONTEXT DEFINITION AND CLASSIFICATION Mobile user context is defined [1] as ”any information that can be used to characterize the situation of the entity, where an entity can be a person, place, physical or computational object”. Schilit et al. [2] claim that important aspects of the context answer the following questions where are you, who are you with, and what resources are near you. Therefore the context should be classified into three categories describing the information related to: • computing environment, e.g. available processors, device capabilities, network capacity, connectivity and computing cost • user environment described with his location infor- mation, social situation and activity, willingness to communicate, preferences etc. • physical environment referring to conditions in the environment, such as lighting, noise level, tempera- ture etc.