Route Recommendations to Business Travelers Exploiting Crowd-Sourced Data

Business travellers are those people who attend work-related meetings and in their few hours of spare time would like to see the best that the host city can offer in terms of cultural activities and sightseeings. In this work we present a complex architecture, consisting of mobile applications and back-end server components, which supports business travelers in recommending possible routes matching their preferences within their timing constraints. The three main contributions are (i) a set of machine learning algorithms that can be used to detect a queuing state of a user with a high degree of accuracy, (ii) how to determine user’s positioning, and (iii) how to practically realize a planner providing a reasonably good enough route plan within a handful of seconds. Preliminary tests demonstrate that the single components of the proposed architecture are feasible and provide good results