NEXYTE product description Page 9
6.2 Explainable ML Models
Machine Learning explainability is about understanding and interpreting the decisions made by
machine learning models. It builds trust and transparency by uncovering the factors that influence
predictions. This helps experts assess reliability, identify biases, and make informed decisions. ML
explainability ensures responsible and ethical use of AI.
To ensure transparency, ML-generated insights and enrichments in NEXYTE are accompanied by
explanatory information which provides users with comprehensive understanding of the models’
behavior. For example, some ML-generated predictions are provided with decision trees that visualize
the model’s decision flow when generating them. Alternatively, NEXYTE can present the impact and
magnitude that each feature had on a model’s prediction.
Visualization of a model's decision flow
Explanation of which features in a risk scoring model led to a high score