Insight
Accuracy is not enough: read a prediction model through its mistakes
A model can achieve a strong overall score while making the mistakes your organization cares about most.
Machine Learning
Systems that learn patterns from data, including training quality, evaluation, changing inputs and limits of prediction.
AI-generated editorial illustration.
Understand Machine Learning
Machine learning uses examples to build systems that identify patterns or make predictions. Here we explain data preparation, training, evaluation and the interpretation of errors. Readers can explore how a model is assessed for a particular task and why changing inputs can make an earlier performance result less informative.
This topic belongs to Artificial Intelligence & Automation. The parent sector collects related coverage across its topics.
Latest Insights
Full archive →News and reference briefs
Full archive →News brief
Machine-learning evaluation needs the cost of an error, not just a score
Prediction quality is a conversation about decisions.