Question

This machine-learning algorithm stores training data for future reference. Therefore, when deployed, the algorithm makes predictions for new data input by merely comparing the stored data samples which are closer to it. The prediction will follow the majority vote for classification or weighted average of neighboring samples for regression. This kind of approach is very flexible as it discards any assumption on the model but it may slow and can overfit to the selected training data samples. What do you categorize the kind of model?

parametric model
nonparametric model
online learning
batch leaming

Answer 1

The kind of model described is a nonparametric model that stores training data for future reference and makes predictions based on the similarity to the stored data.

Step-by-step explanation:

The kind of model described in the question is a nonparametric model. Nonparametric models do not make specific assumptions about the functional form of the relationship between the input and output variables. Instead, they rely on the data to determine the relationship. In this case, the algorithm stores training data and makes predictions based on the input data's similarity to the stored data.