C3 Data Science Methodology

Question 1

Define Methodology

Answer 1

A system of methods used in a particular domain.

Question 2

10 stages of Data Science Methodology?

Answer 2

Get Business understanding.
Determine Analytic approach.
Data requirements.
Data collection.
Data understanding.
Data preparation.
Modeling.
Evaluation.
Deployment.
Feedback.

Question 3

(1.) Business Understanding asks?

Answer 3

What is the problem that you’re trying to solve?

Question 4

An (2.) Analytic Approach asks?

Answer 4

How can I use data to answer the question?

or,

What patterns address the question most effectively?

Question 5

Evaluating (3.) Data Requirements asks?

Answer 5

What data do you need to answer the question?

Question 6

Planning (4.) Data collection asks?

Answer 6

Where will I get the data that I need, and how will I ingress it?

Question 7

(5.) Data Understanding asks?

Answer 7

Does the data I’ve collected represent the problem to be solved?

Question 8

During (6.) Data Preparation ask?

Answer 8

What additional work is required to manipulate and work with the data?

Question 9

When (7.) Modeling, ask?

Answer 9

When you apply your data visualizations, do you see answers that address the business problem?

Question 10

During (8.) Evaluation ask?

Answer 10

Does the data model answer the initial business question or must you adjust the data?

Question 11

During (9.) Deployment ask?

Answer 11

Can you put the model into practice?

Question 12

When seeking (10.) Feedback, ask?

Answer 12

Can I get constructive feedback from the data and the stakeholder to answer the business question?

Question 13

What is a cohort?

Answer 13

A group that shares a common characteristic.

Question 14

What is CRISP-DM?

Answer 14

Cross-Industry Standard Process for Data Mining.

Question 15

Predictive models do what?

Answer 15

Tell us the probability of a future event based on historical data.

Question 16

In broad terms, Descriptive Models do what?

Answer 16

Summarize data, without making predictions

Question 17

What is a Feature?

Answer 17

A characteristic or attribute developed within the data that helps in solving the problem.

Question 18

What are, Pairwise Correlations?

Answer 18

An analysis to determine the relationships and correlations between different variables.

Question 19

What is, Text Analysis?

Answer 19

Analyzing and manipulating textual data, & extracting meaningful information and patterns.

Question 20

Descriptive Analytics tells us?

Answer 20

What happened. Hindsight.

Low value & low difficulty.
Provides information.

Question 21

Diagnostic Analytics tells us?

Answer 21

Why did it happen. Insight.

Middling value & difficulty.
Provides information.

Question 22

Predictive Analytics tells us?

Answer 22

What will happen. Forsight.

High value & high difficulty.
Helps us optimize.

Question 23

Prescriptive Analytics tells us?

Answer 23

How we can make it happen. Greater foresight.

Highest value & difficulty.
Helps us optimize.

Question 24

What does ROC curve stand for?

Answer 24

Receiver Operating Characteristic Curve.

First developed during World War II to detect enemy aircraft on radar.

Question 25

Senesitivity formula?

Answer 25

Sensitivity = TP / P = TP / (TP + FN)

Also known as:
Hit Rate
True Positive Rate
Recall

Question 26

Specificity formula?

Answer 26

Specificity = TN/N = TN / (TN+FP)

Also known as:
Selectivity
True Negative Rate

Question 27

What is PPV?

Answer 27

Positive Predictive Value.

Question 28

What is NPV?

Answer 28

Negative Predictive Value.

Question 29

PPV formula?

Answer 29

TP / (TP+FP) = Positive Predictive Value

Question 29

What is a Confusion Matrix?

Answer 29

A matrix thus:

|TP|FN|
————
|FP|TN|

Question 29

NPV formula?

Answer 29

TN / (TN+FP) = Negative Predictive Value.

Question 30

Type I error?

Answer 30

A False Positive.

Question 31

What is a Type II Error?

Answer 31

A False Negative.

Question 32

List the CRISP-DM steps.

Answer 32

Business Understanding.
Data Understanding.
Data Preparation.
Modeling.
Evaluation.
Deployment.