Visualizations and Exploratory Analysis

Question 1

Think about a database. What is an attribute (or dimension, feature, variable)?

Answer 1

An attribute is a data field, representing a characteristic or a feature of a data object.

e.g., customer_ID, name, address

It is essentially what you see in columns

Question 2

When talking about a database, the rows are data objects and the columns are object’s attributes. True or false?

Answer 2

Very true.

Question 3

Give some examples of the following attribute types:

a. Nominal
b. Binary
c. Ordinal
d. Numeric (quantity, interval, ratio)

Answer 3

Nominal = categories, states, or “names of things”
e.g., hair color, marital status, occupation

Binary

• nominal attribute with only two states (0 and 1)
• symmetric binary = both outcomes are equally important (e.g. gender)
• asymmetric binary = outcomes are not equally important (e.g. medical test, positive or negative)

Ordinal = values have a meaningful order but the magnitude between successive values is unknown
e.g., small medium large

Numeric

• quantity: integer, real number
• interval: measured on a scale of equal-sized units
• ratio: e.g., length, monetary quantities

Question 4

What is the difference between discrete and continuous attributes?

Answer 4

Discrete - has only a finite or countably infinite set of values (e.g., zip codes, profession)

Continuous - has real numbers (e.g., temperature, height, weight), technically they can be unbounded, but in reality not really

Question 5

Mean = ?

Answer 5

Average. you know what it is

Question 6

Median = ?

Answer 6

Middle value if odd number of values, or average of the middle two values otherwise

Question 7

Mode = ?

Answer 7

Value that occurs most frequently in the data

Question 8

In a bell shaped distribution, mean = median = mode. True or false?

Answer 8

True

Question 9

What does a box-plot graph display?

Answer 9

Minimum, Q1 (first quartile), Median, Q3, Maximum

Question 10

What is an outlier?

Answer 10

A point beyond a specified outlier threshold. Can be easily (or not?) spotted in graphs

Question 11

Do you know the properties of a normal distribution curve?

Answer 11

You should. Check online.

Question 12

What is a histogram?

Answer 12

A graph display of tabulated frequencies, shown as bars

Question 13

What is a quantile plot?

Answer 13

A plot that displays all of the data for two variables, allowing the user to assess both overall behavior and unusual occurrences

Question 14

Why is a scatter plot useful?

Answer 14

Provides a first look at bivariate data to see clusters of points, outliers etc.