4. Data Analysis
ROC/AUC for probability of sale within d days using random
forest algorithm.
0.0 0.2 0.4 0.6 0.8 1.0
False positive rate
0.0
0.2
0.4
0.6
0.8
1.0
Truepositiverate
ROC Toyota Camry: Prob. of sale within d days
Mean CV ROC d =3 (AUC = 0.82)
Mean CV ROC d =6 (AUC = 0.82)
Mean CV ROC d =9 (AUC = 0.83)
Mean CV ROC d =12 (AUC = 0.82)
Mean CV ROC d =15 (AUC = 0.83)
4 / 6
5. Data Analysis II
Make suggestions for listing-specific features to improve
likelihood of sale.
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
Price
#Sentences/#Words
Length Description
Lexical Diversity
Odometer
# Pictures
Year
Length Title
Location
Color
Car Type
Phone Yes/No
Map Yes/No
Transmission
Cylinders
Drive
Car Status
Fuel
Feature Importances RF Honda Civic d=4
Feature Importances
5 / 6
6. About me
Ph.D. in Computational Physics
Swiss Federal Institute of
Technology (ETH) Zurich
Postdoc in Computational Biology,
Mount Sinai Hospital, Toronto
Software Developer (Business
Application)
Lecturer (Mathematics),
University of Saskatchewan
6 / 6