How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
x
ŷ
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Local PDP
Sensitivity
analysis
x
ŷ
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Sensitivity
analysis
x
ŷ
Local PDP
ICE plot
Repeat
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Sensitivity
analysis
x
ŷ
Local PDP
ICE plot
Repeat
β
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Local PDP
ICE plot
Contribution (β)
Sensitivity
analysis
Repeat
x
ŷ
β
x
β
Our contribution
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Local PDP
ICE plot
Contribution (β)
LCV plot
Sensitivity
analysis
Sensitivity
analysis
Repeat
x
ŷ
β
x
β
Our contribution
How?
Pre-existing techniques as elementary building blocks.
Prediction (ŷ)
Local PDP
ICE plot
Contribution (β)
LCV plot
GCV plot
Sensitivity
analysis
Sensitivity
analysis
Repeat
Repeat
x
ŷ
β
x
β
Contribution-Value plots
Wine acidity (pH)
Contribution-Value plots
An exemplary data model exploration
Data
- 998 red wines
- 11 features
Model
- Random Forest (100 trees)
- Predict quality: 👍 / 👎
Contribution-Value plots
An exemplary data model exploration
Contribution-Value plots
An exemplary data model exploration
Wine acidity (pH)
Contribution-Value plots
An exemplary data model exploration
??
Contribution-Value plots
Line fading . .
Contribution-Value plots
Conclusion