You are preparing a tabular dataset to train a machine learning model that predicts whether a customer will churn. The dataset includes columns such as 'Monthly_Spend', 'Contract_Length', and 'Customer_Age'. In machine learning terminology, what term is used to describe these individual input variables used to make predictions?

In machine learning, data preparation is a foundational step before training any model. Understanding the vocabulary of datasets is critical. The term feature describes an individual measurable property or characteristic of a phenomenon being observed. In the context of tabular datasets used in machine learning, features correspond to the independent input variables (columns) that the model analyzes to learn patterns and make predictions. For example, to predict housing prices, features might include square footage, number of bedrooms, and zip code. The target label refers to the dependent output variable that the model is trying to predict. In supervised learning, this is the ground-truth label associated with each instance in the training data (e.g., whether a customer churns or the final sale price of a house). Hyperparameters are the configuration settings external to the model that cannot be learned directly from the training data. Examples include the learning rate, the number of trees in a random forest, or the number of hidden layers in a neural network. They must be set before training begins to guide the optimization process. The loss function is a mathematical method used to measure how well a machine learning model's predictions match the actual target labels during training. The goal of optimization is to minimize the output of the loss function. Therefore, the input columns/variables representing the properties used for predicting the outcome are correctly referred to as features.