Data Science Prerequisites - Numpy, Matplotlib, and Pandas in Python - What Is a Feature Vector?

The video tutorial introduces vectors and feature vectors, explaining their significance in machine learning. It discusses different approaches to feature engineering, including using domain knowledge and mathematical techniques like polynomial expansion. The tutorial highlights the power of combining these methods and the role of deep learning in automatic feature extraction.

What is a vector primarily described as in the context of this lesson?

• A single number

• A list of numbers

• A geometric shape

• A mathematical equation

In the example of predicting exam grades, what does the 'D' in an N by D matrix represent?

• Number of samples

• Number of predictions

• Number of vectors

• Number of features

Why is it important for a feature to be predictive?

• To simplify the model

• To make the data set larger

• To help predict the output variable

• To ensure it is visually appealing

Which approach to feature engineering relies on domain knowledge?

• Polynomial expansion

• Using expert knowledge

• Random selection

• Deep learning

What is a key advantage of using mathematical techniques like polynomial expansion in feature engineering?

• They do not require understanding of the data

• They are faster to implement

• They require deep domain knowledge

• They are always more accurate

How do deep neural networks handle feature extraction?

• They require manual feature engineering

• They use only domain knowledge

• They automatically extract features

• They ignore features

What is the typical approach in deep learning regarding feature engineering?

• No feature engineering

• Only domain knowledge

• Manual feature engineering

• Hybrid approach