Microsoft Azure AI Fundamentals (AI-900) Practice Exam

What is the term used to describe the data values that have an impact on the predictions made by a model?

1. Outcomes

2. Features

3. Labels

4. Variables

The correct answer is: Features

The term that refers to the data values impacting the predictions made by a model is "features." In the context of machine learning, features are the individual measurable properties or characteristics of the phenomena being observed. They function as the input variables used by algorithms to learn patterns and make predictions. By thoughtfully selecting and engineering features from raw data, one can significantly influence a model's performance. Good features can help improve accuracy, while irrelevant or poorly chosen features can introduce noise, leading to less reliable predictions. Outcomes refer to the results or outputs predicted by a model rather than inputs. Labels are typically used to denote the actual outcomes associated with the features during the training phase, especially in supervised learning. Variables are a more general term that can refer to any storage location identified by a name, containing data that can change, but in the context of machine learning, it doesn't specifically denote those critical input properties driving model predictions.