Data Science

Question 1

What is precision?

Answer 1

TP/(TP+FP). Proportion of correctly classified positives.

Question 2

What is recall?

Answer 2

TP/(TP+FN). Proportion of correctly classified positives and actual positives

Question 3

What is sensitivity?

Answer 3

Same as recall: (TP/(TP+FN))

Question 4

What is specificity?

Answer 4

TN/(TN+FP). “Negative” variant of recall or sensitivity.

Question 5

What is accuracy?

Answer 5

(TP+TN)/(TP+TN+FP+FN). Proportion of correctly classified samples over all samples.

Question 6

What is the coefficient of determination (r-squared)?

Answer 6

The proportion of the variation in the dependent variable that is predictable or explainable from the independent variable(s). 1-(residual sum of squares/total sum of squares)

Question 7

What is a residual in regression analysis?

Answer 7

The vertical distance between a data point and the regression line

Question 8

What is the total sum of squares?

Answer 8

The sum over all squared differences between the observations and their overal mean

Question 9

What is bootstrapping?

Answer 9

A resampling technique used to estimate the sampling distribution of an estimator by sampling with replacement from the original sample

Question 10

What is sampling with replacement?

Answer 10

Sampling with replacement means that after a unit (e.g., a person, an observation, etc.) is selected from the sample, it is returned to the sample before the next unit is selected. This means that the same unit can be selected more than once in the sampling process

Question 11

What is bagging?

Answer 11

a machine learning ensemble method that stands for “bootstrapped aggregation.” It is a technique used to improve the stability and accuracy of machine learning models by training several models on different subsets of the data and averaging their predictions

Question 12

What is a greedy algorithm?

Answer 12

An algorithmic paradigm that follows the problem-solving heuristic of making the locally optimal choice at each stage with the hope of finding a global optimum

Question 13

What is dynamic programming?

Answer 13

An algorithmic paradigm that solves problems by breaking them down into smaller subproblems, storing the solutions to these subproblems, and then using these stored solutions to solve the larger problem

Question 14

What is bias?

Answer 14

The error that is introduced by simplifying assumptions made by the model

Question 15

What is variance?

Answer 15

The error that is introduced by the sensitivity of the model to small fluctuations in the data

Question 16

In what situation is precision more important than recall?

Answer 16

When a false positive is more acceptable than a false negative

• Classifier to identify spam emails
• Search engine results

Question 17

In what situation is recall more important then precision?

Answer 17

When it’s important to find (all) instances of the positive class.

• Diagnosing a disease
• Detecting fraud

Question 18

What are the four types of analytics technique that are used in organizations?

Answer 18

• Descriptive analytics (what happened?)
• Diagnostic analytics ( why did it happen?)
• Predictive analytics (what will happen?)
• Prescriptive analytics (what should be done?)