Explaining model transactions

The Watson OpenScale service provides explanations for your model transactions to help you understand how predictions are determined.

You can analyze local explanations to understand the impact of factors for specific transactions or analyze global explanations to understand the general factors that impact model outcomes. The type of model that you configure determines the type of explanation that you can use to analyze your transactions.

Watson OpenScale supports explanations for structured, image, and unstructured text models. Structured models can use binary, multiclass, or regression classification problems. Image or unstructured text models can use binary or multiclass classification problems.

When you configure explainability in Watson OpenScale, you can use Local Interpretable Model-Agnostic Explanations (LIME), contrastive explanations, or Shapley Additive explanations (SHAP) to analyze transactions.

LIME identifies which features are most important for a specific data point by analyzing up to 5000 other close-by data points. In an ideal setting, the features with high importance in LIME are the features that are most important for that specific data point.

Contrastive explanations calculate how many values need to change to change the prediction or maintain the same prediction. The factors that need the maximum change are considered more important, so the features with the highest importance in contrastive explanations are the features where the model is least sensitive. For contrastive explanations, Watson OpenScale displays the maximum changes for the same outcome and the minimum changes for a changed outcome. These categories are also known as pertinent positive and pertinent negative values. These values help explain the behavior of the model in the vicinity of the data point for which an explanation is generated.

SHAP is a game-theoretic approach that explains the output of machine learning models. It connects optimal credit allocation with local explanations by using Shapley values and their related extensions. SHAP assigns each model feature an importance value for a particular prediction, which is called a Shapley value. The Shapley value is the average marginal contribution of a feature value across all possible groups of features. The SHAP values of the input features are the sums of the difference between baseline or expected model output and the current model output for the prediction that is being explained. The baseline model output can be based on the summary of the training data or any subset of data that explanations must be generated for. The Shapley values of a set of transactions can be combined to get global explanations that provide an overview of which features of a model are most important.