Which concept refers to splitting data into training and testing sets?

Data partitioning is the fundamental machine learning process of dividing a dataset into two or more subsets. The most common partition is splitting the data into a training set, used to train the model, and a testing (or validation) set, used to evaluate the model's performance on unseen data. This practice is crucial for assessing how well a model generalizes to new data and for preventing overfitting, where a model performs well on training data but poorly on new data.

A. Data drift: This refers to the change in data distribution over time, which can degrade model performance after deployment, not the initial splitting of data.

C. Model deployment: This is the process of integrating a trained model into a production environment to make predictions, which occurs after training and testing.

D. Feature scaling: This is a data preprocessing technique used to normalize the range of features, not to split the dataset itself.

1. Microsoft. (2023, October 24). Split Data module. Azure Machine Learning, Microsoft Learn. Retrieved from https://learn.microsoft.com/en-us/azure/machine-learning/component-reference/split-data. (See the "Overview" section).

2. Ng, A. (n.d.). Machine Learning Yearning, Chapter 5: Basic Error Analysis. Stanford University. Retrieved from https://www.deeplearning.ai/machine-learning-yearning/. (This chapter discusses the importance of splitting data into train/dev/test sets).