This document summarizes research on analyzing urban transportation data to help travelers make more cost-effective decisions. The researcher analyzed anonymous payment data from London's Oyster cards, finding that travelers collectively overspend £200 million per year by not choosing the optimal fares. Machine learning recommender systems were tested on this data and achieved 74-98% accuracy in recommending lower-cost fare options. This research aims to design decision support tools that can save travelers money by better matching their transit needs and patterns to available fare options.
15. Purchase Geography Mobility Flow
45
Zone 1
40
PAYG Zone 2
Travel Cards Zone 3
35
Zone 4
30 Zone 5
Zone 6
25
arrive
20
15
10
5 depart
0
1 2 3 4 5 6 7 8 9
16. high regularity – in movement,
purchases
small increments, short terms
is this ideal?
17. luckily,
computers are good at
counting. let them do it.
idea:
compare what you bought to
what you could have bought
(was it cheaper?).
repeat 300,000 times.
19. using this sample to estimate the entire
city means we overspend by:
£200 million per year
by making the wrong decisions.
20. £200 million per year
by making the wrong decision?
not understanding how we will
need public transport (but..)
failing to match fares with our
needs (but...)
28. further reading:
N. Lathia, L. Capra. Mining Mobility Data to Minimise Travellers'
Spending on Public Transport. In ACM KDD 2011, San Diego, USA.
N. Lathia, J. Froehlich, L. Capra. Mining Public Transport Data for
Personalised Intelligent Transport Systems. In IEEE ICDM 2010,
Sydney, Australia.
N. Lathia and L. Capra. How Smart is Your Smart card? Measuring
Travel Behaviours, Perceptions, and Incentives. In ACM UbiComp
2011, Beijing, China.