Chapter 10

Question 1

What is big data?

Answer 1

Big data refers to a group of data that is too large to be processed through conventional methods, characterized by high volume and velocity of collection, and variety in type and quality.

Question 2

Nominal Variables

Answer 2

A nominal variable’s categories that have no ordering, existing in name only (i.e. ‘grapes’, ‘oranges’, ‘bananas’, etc).

Question 3

Ordinal variables

Answer 3

Ordinal variable categories have a specific ordering (i.e. ‘agree’, ‘neutral’, ‘disagree’)

Question 4

What are the two broad types of variables? What is the difference?

Answer 4

Quantitative: numerical
Categorical: name/category

Question 5

What are the two types of categorical variables?

Answer 5

Nominal: unordered, exist in name only
Ordinal: part of a set order

Question 6

What are the two types of quantitative variables?

Answer 6

Continuous: infinite along a continuum of values; typically real numbers; typically measurements;

Discrete: values are finite within a range, typically integers. Usually represent countable items/groups of items.

Question 7

What is data visualization? Why does it matter?

Answer 7

When data is displayed in a visual format meant to more easily convey information to people, such as a chart or graph.

Data shown in a text-only format often does not convey ideas or information very clearly.

Question 8

What is cardinality?
Why must you consider this?

Answer 8

The number of unique elements in a data set.

Consider cardinality when choosing a visual method by which to represent data, as some methods (such as pie charts) are only suited to low-cardinality, while others (such as charts) may be better for high cardinality.

Question 9

What is an example of a data visualization method suitable for low-cardinality?
What about high-cardinality?

Answer 9

Pie charts are an example of data visualization suitable for low-cardinality.

Scatter plots and histograms are example of visualization suitable for higher-cardinality.

Question 10

What is a good visual method for representing categorical data?

Answer 10

Bar graphs

Question 11

What command would import a module “pandas” module using the alias “pd”?

Answer 11

import pandas as pd

Question 12

What is matplotlib?

What does it replicate?

Answer 12

matplotlib is a module used for plotting data in Python.

It replicates the capabilities of MATLAB, an engineering-oriented programming langauge.

Question 13

This matplotlib function specifies the title of a plot:

Answer 13

plt.title()

Question 14

This matplotlib function specifies the x-axis label of a plot:

Answer 14

plt.xlabel()

Question 15

This matplotlib function specifies the y-axis label of a plot:

Answer 15

plt.y-label()

Question 16

This matplotlib function creates a text label:

Answer 16

plt.text()

Question 17

This matplotlib function creates a text label for a specific data point:

Answer 17

plt.annotate()

Question 18

This matplotlib function creates a legend for the plot:

Answer 18

plt.legend()

Question 19

What is a data frame / DataFrame?

What are it’s 3 components?

Answer 19

A data frame is a two-dimensional tabular data structure with labeled columns and rows, similar to a spreadsheet.

It has 3 components:
-Index
-Columns
-Values

Question 20

What is pandas?

When is this especially useful?

Answer 20

pandas is a Python library that provides tools for working with data frames (reading, writing, subsetting, and reshaping data, etc.)

Can be useful when dealing with a data set that has missing or misaligned data.

Question 21

This dataframe method outputs summary statistics for numerical columns:

Answer 21

dataframe.describe()

Question 22

This dataframe method outputs the first / last 5 rows in the dataframe:

Answer 22

dataframe.head()
dataframe.tail()

Question 23

This dataframe method outputs the minimum / maximum value in a numerical column:

Answer 23

dataframe.min()
dataframe.max()

Question 24

This dataframe method outputs the mean / median value in a numerical column:

Answer 24

dataframe.mean()
dataframe.median()

Question 25

This dataframe method outputs a random row:

Answer 25

dataframe.sample()

Question 26

This dataframe method outputs the standard deviation of values in a numerical column:

Answer 26

dataframe.std()

Question 27

To select a column or columns of a data frame…

Answer 27

…use the command:

data_frame[“column”]

Question 28

What command returns the first
rows of the a data frame?

Answer 28

dataframe[0:10]

Question 29

How would you read a list of tabulated values into a data frame using pandas?

Answer 29

import pandas as pd

ex_data_frame = pd.read_csv(‘ex_data.csv’)

Question 30

What is subsetting?

Answer 30

Subsetting data is the process of retrieving parts of a data frame

Question 31

What is loc() used for?
What is iloc() used for?

Answer 31

loc() is used to select a range of rows and/or a subset of columns (i.e: titanic.loc[0:5,[“pclass”,”age”]])

iloc() is used to select a range of rows and/or columns (i.e: exdatfrm.iloc([0:5, [0:5])

Question 32

What are the two typical forms of data frames?

Answer 32

long form: each column is a variable and each row gives non-repeated data.

wide form: each data variable is in a different column.

Question 33

What is data reshaping?

Answer 33

This is the process of converting a data frame from long form to wide form or vice-versa.

Question 34

What is pivoting?
What is melting?

Answer 34

Pivoting is the process of converting a data frame from long form to wide form.

Melting is the process of converting a data frame from wide form to long form.