NEXYTE product description Page 10
Explanation of threat scoring outcome
By being able to review ML models’ explainable data, as well as user rankings, NEXYTE users are
capable of making better decisions of how and whether to use these models.
6.3 Vetting of Results
Additionally, NEXYTE users can use voting mechanisms (e.g. thumbs up/down) to review and rank a
model’s outcomes or predictions. Reviewing the average ranking can assist them to understand the
reliability and accuracy of a model. In addition, the organization’s data scientists can decide based on
rankings whether a model need to be fine-tuned or trained on additional data. This capability helps
to guard against negative outcomes and biases.