Describe the weights of features used by the model (including visualization that shows the weights of features).

Explanation example

The AI helper made this recommendation based on how important each characteristic is for determining if a plant is safe or poisonous.

Approximate the model to an interpretable decision tree.

Explanation example

This diagram describes a simplified view of how the AI helper makes decisions. It takes branches at decision points to arrive at a conclusion of whether a plant is safe or poisonous.

Approximate the model to a set of rules, e.g. if-then rules.

Explanation example

The AI helper follows a set of rules to determine if a plant is safe or poisonous. Here is a simplified view of the rules the AI helper uses.

01. If the plant is not spotted, then it is poisonous.
02. If the plant is thorny, then it is poisonous.
03. If the plant is small, then it is poisonous.
04. If the plant is white, then it is poisonous.

Show what information the system has access to in order to make decisions.

Explanation example

The AI helper uses color, appearance, texture, and size to identify plants after analyzing 1000 plants of the same breed.

Show what the system is able to do.

Explanation example

The AI helper can help identify if this breed of plant is poisonous or not.

Show how features of the instance contribute to the model’s prediction (including causes in parts of an image or text).

Explanation example

The AI helper calculated the weight of each characteristic to determine if the plant is safe or poisonous.

Describe the rules or a decision-tree path that the instance fits to guarantee the prediction.

Explanation example

This diagram describes a simplified view of how the AI helper decided that this plant is poisonous.

Describe the feature(s) that will change the prediction if perturbed, absent or present.

Explanation example

Texture
If the texture was not thorny, then the plant would have been predicted to be safe.

Provide example(s) similar to the instance and with the same record as the prediction.

Explanation example

Provide example(s) with small differences from the instance but with a different record from the prediction.

Explanation example

Show how the prediction changes corresponding to changes of a feature (often in a visualization format).

Explanation example

See if and how the prediction changes when you change one or more of the characteristics.

Explain how certain the AI is in its prediction, e.g. through category (high, low), numbers, N-best alternatives, or visualizations.

Explanation example