What are your top ten favorite movies of all time? This is a very difficult question. But why? Irmak Sirer explains the challenges of measuring how much we like movies, books, songs, or products; combining insights from diverse sources like the Netflix Prize, Duncan Watts' social experiments, or the beginnings of Facebook. The better we get at measuring and ranking levels of enjoyment, the better we can customize websites, sort search results, find other people with similar tastes, and recommend products, so can we overcome these challenges? Drumroll... Yes, we can.
52. How did they do it?
Before:
Solid assumptions
You have a certain taste.
Your taste dictates a hidden rating for Book of Eli.
When you watch it, this rating is revealed to you.
53. How did they do it?
Before:
Solid assumptions
G
N
O
R
W
You have a certain taste.
Your taste dictates a hidden rating for Book of Eli.
When you watch it, this rating is revealed to you.
54. How did they do it?
After:
Your rating changes with time.
55. How did they do it?
After:
Your rating changes with time.
It depends on...
56. How did they do it?
After:
Your rating changes with time.
It depends on...
how many you rated that day
your average rating for the day
which movies you rated on this day
shown Netflix prediction
57. Trivial: Mean score of everyone
Error:
1.0540 stars
Cinematch
Error:
0.9525 stars
Y. Koren, The BellKor Solution to the Netflix Grand Prize. 2009
58. Trivial: Mean score of everyone
Error:
1.0540 stars
Cinematch
Error:
0.9525 stars
Your time dependent rating tendencies
Y. Koren, The BellKor Solution to the Netflix Grand Prize. 2009
59. Trivial: Mean score of everyone
Error:
1.0540 stars
Cinematch
Error:
0.9525 stars
Your time dependent rating tendencies
Error:
0.9278 stars
Y. Koren, The BellKor Solution to the Netflix Grand Prize. 2009
60. Trivial: Mean score of everyone
Error:
1.0540 stars
Cinematch
Error:
0.9525 stars
12.0%
Your time dependent rating tendencies
Error:
0.9278 stars
Y. Koren, The BellKor Solution to the Netflix Grand Prize. 2009
61. Trivial: Mean score of everyone
Error:
1.0540 stars
Cinematch
Error:
0.9525 stars
12.0%
Your time dependent rating tendencies
Error:
0.9278 stars
without looking at which movies you like/hate!
Y. Koren, The BellKor Solution to the Netflix Grand Prize. 2009
63. What does this suggest?
We cannot compare a movie with all others we've seen.
64. What does this suggest?
We cannot compare a movie with all others we've seen.
We compare it to a limited set.
65. What does this suggest?
We cannot compare a movie with all others we've seen.
We compare it to a limited set.
Liking (real time & remembered) depends on time and mood.
66. What does this suggest?
We cannot compare a movie with all others we've seen.
We compare it to a limited set.
Liking (real time & remembered) depends on time and mood.
Other people's opinions affect our own (followers / hipsters)
67. What does this suggest?
We cannot compare Book of Eli with all movies we've seen.
We compare it to a limited set.
Liking (real time & remembered) depends on time and mood.
Other people's opinions affect our own (followers / hipsters)
69. An experiment
Same website: Music download and rating
M.J. Salganik, P.S. Dodds, D.J. Watts. Science, 311:854-856, 2006
70. An experiment
Music Lab: A website for downloading music
Alternative A:
Other people's ratings invisible
71. An experiment
Music Lab: A website for downloading music
Alternative A:
Other people's ratings invisible
More or less equal ratings
72. An experiment
Music Lab: A website for downloading music
Alternative A:
Other people's ratings invisible
Alternative B:
All ratings visible
More or less equal ratings
73. An experiment
Music Lab: A website for downloading music
Alternative A:
Other people's ratings invisible
Alternative B:
All ratings visible
More or less equal ratings
Several songs snowball in popularity
74. An experiment
Music Lab: A website for downloading music
Alternative A:
Other people's ratings invisible
Alternative B:
All ratings visible
More or less equal ratings
Several songs snowball in popularity
It's different songs for each trial
76. Problems with rating movies
We cannot compare a movie with all others we've seen.
We compare it to a limited set.
Liking (real time & remembered) depends on time and mood.
Other people's opinions affect our own.
77. Degree of liking is
sensitive and vague
Amazing!
Tuesday 3am
Total
garbage
Sunday 12pm
78. Degree of liking is
sensitive and vague
Liking (real time & remembered) depends on time and mood.
Other people's opinions affect our own.
79. Degree of liking is
sensitive and vague
Dependent on many other
environmental factors
besides our taste
80. Degree of liking is
sensitive and vague
We cannot compare a movie with all others we've seen.
We compare it to a limited set.
81. Degree of liking is
sensitive and vague
Difficult to describe
accurately and consistently
with a number
95. Trying to rate Star Wars
1
Map enjoyment
to a specific scale
96. Trying to rate Star Wars
1
Map enjoyment
to a specific scale
97. Trying to rate Star Wars
1
Map enjoyment
to a specific scale
98. Trying to rate Star Wars
2
rating
ose corresponding
cho
king
for this degree of li
99. Trying to rate Star Wars
But we cannot keep
this entire history of
enjoyment in mind
100. Trying to rate Star Wars
But we cannot keep
this entire history of
enjoyment in mind
We fuzzily remember
a small subset
101. Trying to rate Star Wars
But we cannot keep
this entire history of
enjoyment in mind
We fuzzily remember
a small subset
We map based on this subset
102. Trying to rate Star Wars
But we cannot keep
this entire history of
enjoyment in mind
We fuzzily remember
a small subset
We map based on this subset
176. How did they do it?
After:
A small, constant increase
in uncertainty before each
comparison
Probability
Your rating changes with time.
3.5
4
4.5
5
uncertainty
177. Degree of liking is
sensitive and vague
Great! We have a system!
178.
179. How many is too many?
I don’t want to
spend too much
time on this
200. Quantifying human reactions are hard
books
celebrities
songs
tv shows
food
importance of issues
politicans
what to spend ‘fun’ budget on
products
teams in different sports
201. Degree of liking is
sensitive and vague
Amazing!
Tuesday 3am
Total
garbage
Sunday 12pm
203. Quantifying reactions is very useful
customized websites
sorting search results
recommendations
connecting with other people of similar tastes
identifying meaningful groups of
similar products / people
understanding your own preferences
205. Quantifying human reactions are hard
Start with a rating,
pose the correct comparisons
Every decision gets us closer
206. Degree of liking is
sensitive and vague
Amazing!
Tuesday 3am
Total
garbage
Sunday 12pm
207. Many comparisons for a movie
over different days
averages out mood and other factors
208. Many comparisons for a movie
over different days
averages out mood and other factors
We can’t do much about social influence,
but we should just accept that
as natural part of how much we like things
209. Degree of liking is
sensitive and vague
Amazing!
Tuesday 3am
Total
garbage
Sunday 12pm
210.
211. A great way of collecting desired data
is to make it fun