The recent availability of large amount of mobility data has fostered many research efforts to improve mobility prediction. Lots of these studies are focused on learning the impact of influencing factors on traffic, such as rush hour or accidents. Nevertheless, only very few have investigated the impact of Planned Special Events (PSEs), such as concerts, soccer games, etc., despite their well-known influence on traffic. In this paper we present an automatic solution to model the impact of PSEs on traffic around the venue of the events. In particular, we answer the question of "which road segments are affected by PSEs?" by identifying which roads show an event specific behavior that can identify the happening of a PSE reliably. For that, we propose a solution based on an Artificial Neural Network (ANN) classifier that is trained on traffic data on event and non-event days for each road. The proposed approach has been evaluated on two different venues in Germany with a leave-one-out cross-validation performed on all the soccer matches played in those locations during the season 2013/14 of the German First League. Results show that the approach can reliably identify road segments affected by PSEs, with an F-Measure up to 0.97.
Stuck around the Stadium? An Approach to Identify Road Segments Affected by Planned Special Events / Kwoczek, Simon; DI MARTINO, Sergio; Nejdl, Wolfgang. - (2015), pp. 1255-1260. (Intervento presentato al convegno 18th IEEE International Conference on Intelligent Transportation Systems (ITSC 2015) tenutosi a Gran Canaria, Spain nel September 15-18, 2015) [10.1109/ITSC.2015.206].
Stuck around the Stadium? An Approach to Identify Road Segments Affected by Planned Special Events
DI MARTINO, SERGIO;
2015
Abstract
The recent availability of large amount of mobility data has fostered many research efforts to improve mobility prediction. Lots of these studies are focused on learning the impact of influencing factors on traffic, such as rush hour or accidents. Nevertheless, only very few have investigated the impact of Planned Special Events (PSEs), such as concerts, soccer games, etc., despite their well-known influence on traffic. In this paper we present an automatic solution to model the impact of PSEs on traffic around the venue of the events. In particular, we answer the question of "which road segments are affected by PSEs?" by identifying which roads show an event specific behavior that can identify the happening of a PSE reliably. For that, we propose a solution based on an Artificial Neural Network (ANN) classifier that is trained on traffic data on event and non-event days for each road. The proposed approach has been evaluated on two different venues in Germany with a leave-one-out cross-validation performed on all the soccer matches played in those locations during the season 2013/14 of the German First League. Results show that the approach can reliably identify road segments affected by PSEs, with an F-Measure up to 0.97.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
