Practice Library
All MCQs
Browse exam-wise, subject-wise, and country-wise MCQs with explanations.
Choose an option to check your answer.
A.
Creating new attributes from existing data
B.
Deleting the target variable
C.
Changing row order only
D.
Selecting a random algorithm
Show Answer
Correct Answer: A. Creating new attributes from existing data
Explanation:
Constructed features may expose useful relationships not directly present in raw variables.
Choose an option to check your answer.
A.
The variables are always independent
B.
The variables have zero variance
C.
The data contain no noise
D.
There is no linear association, though a nonlinear relationship may exist
Show Answer
Correct Answer: D. There is no linear association, though a nonlinear relationship may exist
Explanation:
Pearson correlation measures linear dependence only.
Choose an option to check your answer.
A.
Correlation always proves causation
B.
Correlation ignores direction
C.
Correlation is standardized and unitless
D.
Correlation can only be positive
Show Answer
Correct Answer: C. Correlation is standardized and unitless
Explanation:
Correlation scales covariance by the variables' standard deviations.
Choose an option to check your answer.
A.
The number of categories in one variable
B.
Whether two variables tend to vary together
C.
The probability of a class label
D.
The amount of missing data
Show Answer
Correct Answer: B. Whether two variables tend to vary together
Explanation:
Positive covariance means variables tend to rise together, while negative covariance indicates opposite movement.
Choose an option to check your answer.
A.
As the average squared deviation from the mean
B.
As the difference between maximum and minimum
C.
As the median of all values
D.
As the most frequent value
Show Answer
Correct Answer: A. As the average squared deviation from the mean
Explanation:
Squaring prevents deviations of opposite signs from cancelling.
Choose an option to check your answer.
A.
The full range
B.
The variance
C.
The standard deviation
D.
The interquartile range
Show Answer
Correct Answer: D. The interquartile range
Explanation:
The IQR depends on the first and third quartiles.
Choose an option to check your answer.
A.
The median
B.
The interquartile range
C.
The arithmetic mean
D.
The mode
Show Answer
Correct Answer: C. The arithmetic mean
Explanation:
The mean uses the magnitude of every observation.
Choose an option to check your answer.
A.
The median observation
B.
An observation that differs markedly from most other observations
C.
A missing column name
D.
A feature with two categories
Show Answer
Correct Answer: B. An observation that differs markedly from most other observations
Explanation:
Outliers may reflect error, rare events, or legitimate extremes.
Choose an option to check your answer.
A.
Replacing values in each bin with that bin's mean
B.
Deleting all observations outside the bin
C.
Sorting bins by variance
D.
Using the global maximum for every value
Show Answer
Correct Answer: A. Replacing values in each bin with that bin's mean
Explanation:
This method reduces local noise by representing nearby values with a common summary.
Choose an option to check your answer.
A.
Using identical numeric widths
B.
Assigning every value to one bin
C.
Making all bin means equal
D.
Creating intervals with approximately equal numbers of observations
Show Answer
Correct Answer: D. Creating intervals with approximately equal numbers of observations
Explanation:
Quantile-based bins balance counts rather than widths.
Choose an option to check your answer.
A.
Creating bins with equal numbers of observations
B.
Using one bin per unique value
C.
Dividing the numeric range into intervals of the same width
D.
Sorting categories alphabetically
Show Answer
Correct Answer: C. Dividing the numeric range into intervals of the same width
Explanation:
Equal-width binning partitions the value range uniformly.
Choose an option to check your answer.
A.
Combining two databases
B.
Converting a continuous attribute into intervals or categories
C.
Removing all categories
D.
Estimating class probabilities
Show Answer
Correct Answer: B. Converting a continuous attribute into intervals or categories
Explanation:
Discretization replaces numeric ranges with bins such as low, medium, and high.
