15. payment vs. reference
compared to 3,379,570 transactions
(5% data set):
49.8% of transactions should fall into
this category.
16. data vs. behaviour & perception
+ estimating travel time > estimating travel needs
+ users inaccurately estimate payments
+ high usage of modalities that are “never used”
+ high proportion of travel to “non typical” stations
21. mobility vs. student discounts
“... as a student with a flexible
timetable, it isn't always obvious
whether a travel card is a better deal
than pay as you go...”
22. mobility vs. daily capping
91,391,155 pay as you go trips
split into capped vs. non-capped
(free) (non-free)
23. mobility vs. daily capping
not-free: 54.83% trips by bus
free: 73.53% trips by bus
24. the oyster card
a tool to evaluate policy?
a sensor to test incentives?
25.
26. N. Lathia, J. Froehlich, L. Capra. Mining Public Transport
Usage for Personalised Intelligent Transport Systems. In
IEEE ICDM 2010, Sydney, Australia, December 2010.
N. Lathia, L. Capra. Mining Mobility Data to Minimise
Travellers' Spending on Public Transport. In ACM KDD 2011,
San Diego, USA, August 2011.
27. how smart is your smart card?
measuring travel behaviours, perceptions, and incentives
@neal_lathia
dept. of computer science
university college london
ubicomp 2011