Feature Engineering And EDA

What is the primary goal of EDA?

• Building machine learning models

• Cleaning data for deployment

• Understanding the data's characteristics and patterns

• Storing data in databases

Which Python library is most commonly used for EDA visualization?

• NumPy

• Pandas

• Matplotlib

• TensorFlow

In EDA, the term "outlier" refers to:

• A missing value in a dataset

• A data point that is significantly different from others

• A feature with too many categories

• A value with no label

Which function in Pandas is used to get the first few rows of a DataFrame?

• head()

• first()

• top()

• sample()

Which plot is best for understanding the distribution of a numerical feature?

• Line plot

• Histogram

• Scatter plot

• Bar chart

Missing values can be handled by:

• Dropping them

• Filling with mean/median/mode

• Predicting them with other data

• All of the above

Which method can be used to detect correlation between numerical variables?

• value_counts()

• corr()

• info()

• describe()

One-hot encoding is mainly used for:

• Numerical features

• Categorical features

• Text data

• Image data

Which of the following is an example of a derived feature?

• The original age column in dataset

• Age group created from age column

• Missing value count in a dataset

• Original height column

Scaling features is important because:

• It reduces missing values

• It makes features comparable in magnitude

• It changes categorical features to numbers

• It removes duplicates