Please enable JavaScript.
Coggle requires JavaScript to display documents.
CHAPTER 19: INTERPRET MODEL (RELATIONSHIP FEATURE CAN HAVE WITH A TARGET…
CHAPTER 19: INTERPRET MODEL
RELATIONSHIP FEATURE CAN HAVE WITH A TARGET
overall impact of a feature without consideration of the impact of other features
treats each feature as a standalone effect on the target
The overall impact of a feature adjusted for the impact of other features.
extent to which the model with the randomly shuffled feature does worse than the original model is assigned as the feature’s value
The directional impact of a feature
whether the presence of a value helps the model by assisting it in predicting
red bars indicate feature values that make positive cases
blue bars indicate feature values that make negative cases
The partial impact of a feature
Model X-Ray - list of features ranked by their influence on the target as denoted by the size of the green line under each feature name in the left pane
partial dependence plot shows the marginal effect of a value when all other features are constant.
word cloud represents the words that have the highest coefficients
intensity of colors indicates the size of their coefficient
analyze large terms with vivid colors
stop word is a common word generally assumed to have little value
of, the
Hotspots
shows the most relevant (up to four) combinations of features and their effect on the target
deeper the tone of the blue and the red, the more of an impact
Reason Codes
pluses or minuses representing the importance of the feature and its value
strong (+++/---)
medium (++/--)
weak (+/-)
left and right thresholds allow for the specification of the probability cutoffs to be used in this view
evaluations of why a prediction was set as the given probability for that case