Roadie: Mobile Semantic Tourism Routes

Roadie is a mobile-based application with two main goals: 1) to provide accurate recommendations of city itineraries enriched with thematic suggestions based on user profiling, 2) to let the user manually create and edit his/her personalised tours through the city.

Roadie has been developed in Java as a native mobile application using the Android SDK. The backend is written in PHP. All the data are stored in a MySQL database. The system is composed by four core modules which are respectively in charge of: 1) grabbing POIs, venues and events, 2) categorising data in realtime, 3) profiling users, 4) recommending routes.

Roadie builds a model of the user profile of interests exploiting user information from the Facebook Graph API through Facebook Login.

Roadie uses two types of geo-localised data: static data, such as POIs and venues, and dynamic data, mainly constituted by events, which change continuously in time. POIs are city attractions retrieved querying the MediaWiki API. GeoNames API is exploited in order to get POI’s latitude and longitude.

Venues are places in the city which can offer visitors commodities, leisure and entertainment services, such as shops, restaurants, nightclubs etc. These data are obtained using Foursquare API. Roadie uses events to enrich the visitor experience in the city. Events are characterised by having an exact start time and duration and are retrieved daily by the grabbing module from two sources: the OpenData (for this demo we used data published by the municipality of Florence, IT), and the Eventful API.

Roadie exploits also physical sensors available in mobile devices in order to observe user behaviours and to refine his/her profile model.

The interface is composed by four main views: 1) recommended routes, 2) route creation, 3) profile and 4) saved routes. The profile view shows demographic data and all the automatically detected interests. These can be edited adding or removing items from the system taxonomy. The recommended routes view provides a scrollable list by which the user can select different thematic itineraries on the basis of his/her interests.

A demo of Roadie application will be presented at the 2015 IEEE International Conference on Multimedia and Expo (ICME 2015), June 29 – July 3, 2015 in Torino, Italy.


Roadie started as a project developed by the students Andrea Barillari and Roberto Medico as final work for the Bachelor of Art at the Faculty of Engineering of University of Florence under the supervision of prof. Alberto Del Bimbo, and his assistants Andrea Ferracani and Daniele Pezzatini.

The work was born from an idea by Andrea Ferracani and Daniele Pezzatini, researchers at MICC and prof. Alberto Del Bimbo collaborators.

About Andrea Ferracani

I'm an interaction designer and an intelligent web applications developer. My research interests focus on machine learning, collective intelligence, rich internet applications, social networks analysis and the semantic web.

Leave a Reply

Your email address will not be published. Required fields are marked *