A Framework for Ridesharing Recommendation Services
A variety of existing ride-on-demand systems support rideshare function besides other functions like traditional taxi. However, many problems have not been solved. First, drivers have to offer their trips and passengers input their request to search for their drivers through a website by smartphone to find a possible match of the trip. Rideshare function of these systems is still limited. Existing systems also fail to provide convenient and flexible ridesharing services for especially regular users with frequent routes. Many drivers and passengers have the same travel demand but have to send the ride requests every day. Last but not least, when people visit a place they often do some specific activity there, for example, eating at a restaurant, and sometimes they do not mind to change to another place where they can do the same activity provided that no additional travel cost and time are incurred. Therefore, to construct proactive real-time ridesharing services, we need to solve all of those problems. This paper focuses on designing a framework for ridesharing and location-based services with the exploitation of knowledge discovered by spatiotemporal data mining techniques. Users can send a ride request anytime. Depending on the time the user needs a ride as well as his activity at the destination, his request can be executed immediately or procrastinated to construct an optimal rideshare and possibly suggest a location for his demanded activity so that the ride fare is lowest.
KeywordsActivity analysis Ridesharing Point of interest
- 1.Financial Times: https://www.ft.com/content/96608536-4204-11e7-9d56-25f963e998b2?mhq5j=e1 (2017). Accessed June 2017
- 2.Nechita, E., Crişan, G.C., Obreja, S.M., Damian, C.S.: Intelligent carpooling system: a case study for Bacău metropolitan area. In: New Approaches in Intelligent Control, pp. 43–72. Springer International Publishing Switzerland (2016)Google Scholar
- 3.Lim, J.H., Chan, J., Karunasekera, S., Leckie, C.: Personalized itinerary recommendation with queuing time awareness. In: The International Conference of SIGIR, pp. 325–334 (2017)Google Scholar
- 4.Furletti, B., Cintia, P., Renso, C.: Inferring human activities from GPS tracks. In: UrbComp 2013 (2013)Google Scholar
- 6.Kalanick, T., Camp, G.: Uber. https://www.uber.com/ (2015). Accessed 30 July 2015