FREQUENTLY ASKED INTERVIEW QUESTIONS

A feature is one piece of information that a Machine Learning model uses to make a prediction.

Example: To predict the price of a house, features can be its area, number of rooms, and location.

A test dataset is new data that the model has not seen before. It is used at the end to check how well the model works. It is like a final exam.

Example: 2,000 new emails can be used to check how well the spam filter works.

A validation dataset is new data used while making a model. It helps us check if the model is learning well and needs any changes. It is like a practice test.

Example: 2,000 new emails can be used to check if the spam filter is learning well before it is ready.

A training dataset is data given to a Machine Learning model so it can learn. It is like a student learning from a book.

Example: 10,000 emails marked "spam" or "not spam" can be used to teach a spam filter.

A Machine Learning model is a program that learns from data. It uses what it learns to make a prediction or decision.

Example: A spam filter learns from old emails and decides if a new email is spam or not.