15. 3 match sentiment with (census) deprivation
2 classify sentiment of profiles
1 collect profiles & geo-reference them
16. 250K profiles in London (31.5M tweets)
3 seeds: newspaper accounts
1 collect profiles & geo-reference them
1,323 in London neighborhoods 573 in 51 neighborhoods
21. read profiles & define topics
create virtual bins (latent topics)
assign words to a bin (@ random)
for each bin:
select pair of words
if co-occur more than chance:
keep them in the bin
else:
put them into another bin (@ random)
Facebook
Twitter
22. read profiles & define topics
create virtual bins (latent topics)
assign words to a bin (@ random)
for each bin:
select pair of words
if co-occur more than chance:
keep them in the bin
else:
put them into another bin (@ random)
Facebook
Twitter
social
econometrics
23. read profiles & define topics
create virtual bins (latent topics)
assign words to a bin (@ random)
for each bin:
select pair of words
if co-occur more than chance:
keep them in the bin
else:
put them into another bin (@ random)
Facebook
Twitter
social
econometrics
52. (a) (b) (c) (d)
(e) (f) (g) (h)
Figure 5: Visual Wordsfor Beauty (top row) and Ugly (bottom row).
and guardrails. The red dots on top-ranked pictures
hose on bottom-ranked ones mean two different things
ormer reflect positive(e.g., happy) visual words, while
Picture Quality Bias. Photos might not necessarily
what they aresupposed to show (representativeurban s
in each neighborhood), and somepictures might beof
53. Research?
This work is at intersection of two emerging fields:
a) computational aesthetic
b) computational geo-cultural modeling
we use the gravity model to study flows of passengers on London’s rail system. extensive network. 588 stations. click for oyster card. oyster cards record the point and time of entry… and link all journeys to a user id, so we can analyse individual travel patterns. stats – 1 month = 77 million journeys by 5 million users between 588 stations.
the world is currently undergoing a massive influx of people into cities. which means that (next slide)
the world is currently undergoing a massive influx of people into cities. which means that (next slide)
the world is currently undergoing a massive influx of people into cities. which means that (next slide)
in particular, transport analysts and planners need to be able to understand and predict passenger flow so far there has been little work on modeling flows on public transport systems in an urban environment, or at the intra- city level. one reason for this may be lack of data, but with adoption of AFC in cities all over the world, this data is now readily available.