Features and Labels in Machine Learning Quiz

Quiz covering Core Machine Learning Concepts

Questions in this quiz

A preview of the 5 questions covered. Start the quiz above to answer them, check your score, and read the explanations.

1. 1

   In a dataset used for predicting house prices, what would be considered the 'label'?

2. 2

   When building a model to classify emails as spam or not spam, which of the following would be a 'feature'?

3. 3

   A dataset contains information about customer demographics and their purchase history. You want to predict if a customer will click on an online advertisement. Which of the following is LEAST likely to be a useful 'feature' for this prediction?

4. 4

   You are working with a dataset to predict the outcome of a soccer match (Win, Lose, Draw). The features include team rankings, player statistics, and historical match results. If you decide to encode the 'home team advantage' as a binary feature (1 if playing at home, 0 if not), this is an example of:

5. 5

   Consider a scenario where you are training a sentiment analysis model on customer reviews. The reviews are labeled as 'positive', 'negative', or 'neutral'. You extract features like the frequency of positive/negative words and the presence of emojis. After training, the model consistently predicts 'neutral' for all reviews, regardless of the actual label. What is the MOST likely cause of this issue, assuming the features were extracted correctly?