Location and Personalisation: Delivering Online and Mobility Services

Information which has been identified as timely and relevant needs to be delivered to users via an appropriate channel. Understanding the user's current activity is crucial in deciding whether a piece of information is relevant to the user, and hence whether the user should be informed about the information.
A classic case of not delivering relevance is the search engine. With the current generation of search engines:
two users with different backgrounds and different interests will get the same results to an identical query;
a user will (usually) get the same results to identical queries in the morning and the afternoon, even though they are working on completely different projects moreover in the evening at home, the same results will be returned, even though the user is in 'leisure' mode;
on revisiting a search engine query, the user will (again usually) be presented with identical results, even though the user has already investigated many of the recommendations.
As an example, consider multiple users searching the Internet for the word 'soap' using one of the many available search engines.
The Google search engine [27] returns over 3.6 million hits for the query 'soap'. But as a homonym, the word will have completely different meanings to the computer software engineer and the beautician. The computer software engineer, having spent the last hour browsing Web information about XML will no doubt be interested in the Google recommendations related to SOAP in the context of XML protocols,...