Chapters 3 - Exploring Data

Question 1

What is data exploration?

Answer 1

A preliminary investigation of the data in order to better understand its specific characteristics. Helps with selecting appropriate preprocessing and data analysis techniques.

Question 2

What are summary statistics?

Answer 2

Quantities such as the mean and standard deviation that capture various characteristics of a potentially large set of values with a single number or a small set of numbers.

Question 3

Define frequency of a data set

Answer 3

frequency(vi) = number of objects with attribute value vi / number of objects eg. 1 2 3 1 1 frequency of 1 is 3/6 or 0.5

Question 4

Define mode of a data set

Answer 4

The mode of a categorical attribute is one that appears most frequently (has the highest frequency)

Question 5

What is a percentile?

Answer 5

The value below which a percentage of data falls eg 80% of people are shorter than you That means you are at the 80th percentile.

Question 6

What do the mean and median measure?

Answer 6

The location of a set of values

Question 7

Define mean

Answer 7

The mean is the average of a set of numbers

Question 8

How is the mean of a set of numbers computed?

Answer 8

Add the values of the numbers and divide by the total number of numbers in the set

Question 9

What is the median?

Answer 9

The median is the middle number in a set of values when arranged from hightest to lowest.

if the set has an even number of values, than the median is the average of the middle two numbers

Question 10

What is a trimmed mean?

Answer 10

A percentage of the beggining and end of the data is thrown out.

This is because the mean is sensitive to outliers.

Question 11

If the distribution of values is skeweked in a data set is the mean a good indivator of the middle of a set of values?

Answer 11

No. In this case, the media is a better indicator of the middle.

Question 12

What do range and variance measure in a data set?

Answer 12

the spread of a set of values

indicate if values are wideley spread out or relatively concentrated around a single point

Question 13

How is the range of a data set calculated?

Answer 13

the largest value subtract the smallest value

Question 14

Why is variance a preffered measure of spread over range?

Answer 14

The range can be misleading if most of the values are concentrated in a narrow band of values, but there are also a relatively small number of mroe extreme values.

Question 15

What symbol is used for variance?

Answer 15

s²_x

Question 16

What is variance?

Answer 16

The average of the squared differences from the mean.

Question 17

How is variance calculated?

Answer 17

To calculate the variance follow these steps:

Work out the Mean (the simple average of the numbers)

Then for each number: subtract the Mean and square the result (the squared difference).

Then work out the average of those squared differences. (

Question 18

What is the standard deviation?

Answer 18

a measure of how spread out the data is

Its symbol is σ (the greek letter sigma)

the formula is the square root of the vaThe formula is easy: it is the square rooriance

Question 19

What can be the issue with using standard devication on data?

Answer 19

The mean can be distorted by outliers, and since the variance is computed using the mean, it is also sensitive to outliers.

Question 20

What is multivariate data?

Answer 20

Measures of location for data that consists of several attrivutes

Question 21

How can measures of location be obtained for data that consists of several attributtes (multivariate data)

Answer 21

Compute the mean or median separately for each attribute.

Question 22

What is the covariance of two attributes?

Answer 22

A measure of the degree to which two attributes vary together and depends on the magnitude of the variables.

A value of 0 indicates that two attributes do not have a (linear) relatioinship.

Question 23

Is it possobile to judge the degree of relationship between two variables by looking only at the value of the covariance?

Answer 23

No, correlatin is preferred to covariance.

Question 24

What does the skewness of a set of values measure?

Answer 24

the degree to which the values are symetrically distributed around the mean.

Question 25

What is usual data mining?

Answer 25

The use of visualization techniques in data mining

Question 26

what is selection (visualization)

Answer 26

the elimination or de-emphasis of certain objects and attributes

there is not a satisfactory or general approach to represent data with many attributes (graphs has 2 or 3 dimensions)

Question 27

What are stem and leaf plots useful for?

Answer 27

Provide insight into the distribution of one-dmensional integer or continous data.

Question 28

What is the following type of visual data representation?

Answer 28

Stem and leaf plot

Question 29

What is a histogram?

Answer 29

a plot that displays the distribution of values for attributes by dividing the possible valuesi nto bins and showing the number of ojbects that fall into each bin.

Question 30

What is a pareto histogram?

Answer 30

A normal histogram, except that categories are sorted by count so that the count is decreasing from left to right.

Question 31

What are box plots?

Answer 31

another method for showing the distrubution of the values of a single numerical attribute.

Question 32

What is the following visual representation of data?

Answer 32

scatter plot

Question 33

What is the following visual representation of data?

Answer 33

Contour plot

Question 34

What is the following visual representation of data?

Answer 34

surface plot

Question 35

What type of visual representation is the following?

Answer 35

vector field plot

a characteristic may have both a magnitude and direction

Question 36

What type of visual representation is the following?

Answer 36

Data matrix

image can be regarded as a rectangular array of pixes

each pixel is characterized byits color and brightness

Question 37

What type of visual representation of data is the following?

Answer 37

parallel coordinates plot

have one coordinate axis for each attribute, but different ax4es are parallelto one other instead of perpendicular

the object is represented as a line instead of a point

the value of each object is mapped to a point on the coordinate axies associated with that attribute

Question 38

What type of visual representation is the following?

Answer 38

Chernoff face

each attribute is associated with a specific feature of a face

each attribute value is used to determine the way tha the facial feature is expressed

Question 39

Define data cube

Answer 39

a multidimensional representation of the data, together with all possible totals (aggregates)

Question 40

Define cross tabulation

Answer 40

Cross-tabluation is about taking two variables and tabulating the results of one variable against the other variable.

An example would be the cross-tabluation of course performance against mode of study:

Question 41

Define pivoting

Answer 41

refers to aggregating over all dimensions except two

the result is a two-diensional cross tabulation with the two specificed dimensions as the only remaining dimensions

Question 42

Define slicing and dicing

Answer 42

slicing: selecting a group of cells from an entire multidimensional array by sepcifying specific values for one or more dimensions.
dicing: selecting a subset of cells by sepcifying a range of attribute value (this is the equivalent of defining a subarray from the complete array).

Question 43

Define “roll up” given a multi-dimensional array of sales data

Answer 43

aggregating the sales across all the dates in a month

Question 44

Define “drill down” given a multi-dimensional array of sales data

Answer 44

splitting monthly sales totals into daily sales totals