Data Mining

2.2 Descriptive Data Summarization

Fall 2008

Instructor: Dr. Masoud Yaghini

Descriptive Data Summarization
 Descriptive Data Summarization

Motivation
– To better understand the data
– To highlight which data values should be treated as noise or
outliers.

Data characteristics
– Measures of central tendency
 Mean, median, mode, and midrange
– Measures of data dispersion
 Rang, quartiles, interquartile range (IQR), and variance

Descriptive Data Summarization

 Outline

Measuring the Central Tendency

Measuring the Dispersion of Data

Graphic Displays of Basic Descriptive Data
Summaries

References

Descriptive Data Summarization

 Measuring the Central Tendency

Descriptive Data Summarization

 Measuring the Central Tendency

In this section, we look at various ways to measure the

central tendency of data, include:
– Mean
– Weighted mean
– Trimmed mean
– Median
– Mode
– Midrange

Descriptive Data Summarization

 Mean

Mean: The most common and most effective numerical measure

of the “center” of a set of data is the (arithmetic) mean. (sample
vs. population)
1 n
x = ∑ xi µ= ∑ x
n i =1 N

Weighted mean: Sometimes, each value in a set may be
associated with a weight, the weights reflect the significance,
importance, or occurrence frequency attached to their respective
values.
n

∑wx
i =1
i i
x = n

∑w
i =1
i

Descriptive Data Summarization

 Trimmed mean

Trimmed mean
– A major problem with the mean is its sensitivity to extreme
(e.g., outlier) values.
– Even a small number of extreme values can corrupt the
mean.
– the trimmed mean is the mean obtained after cutting off
values at the high and low extremes.
– For example, we can sort the values and remove the top and
bottom 2% before computing the mean.
– We should avoid trimming too large a portion (such as
20%) at both ends as this can result in the loss of valuable
information.

Descriptive Data Summarization

 Median

Suppose that a given data set of N distinct values is

sorted in numerical order.

The median is the middle value if odd number of
values, or average of the middle two values otherwise

For skewed (asymmetric) data, a better measure of the
center of data is the median.

Descriptive Data Summarization

 Mode & Midrange

Mode is the another measure of central tendency

– The mode for a set of data is the value that occurs most
frequently in the set.
– If each data value occurs only once, then there is no mode.

The midrange can also be used to assess the central
tendency of a data set
– It is the average of the largest and smallest values in the set.

Descriptive Data Summarization

 Mean, Median, and Mode

Mean, median, and mode of symmetric versus positively and

negatively skewed data.

Positively skewed, where the mode is smaller than the median

(b), and negatively skewed, where the mode is greater than the
median (c).

Descriptive Data Summarization

 Measuring the Dispersion of Data

Descriptive Data Summarization

 Measuring the Dispersion of Data

The degree to which numerical data tend to spread is

called the dispersion, or variance of the data.

The most common measures of data dispersion are:
– Range
– Five-number summary (based on quartiles)
– Interquartile range (IQR)
– Standard deviation

Descriptive Data Summarization

 Range

Range: difference between highest and lowest

observed values

Descriptive Data Summarization

 Inter-Quartile Range

Inter-quartile range (IQR): IQR = Q3 – Q1

– IQR is a simple measure of spread that gives the range
covered by the middle half of the data

Quartiles:
– First quartile (Q1): The first quartile is the value, where
25% of the values are smaller than Q1 and 75% are larger.
– Third quartile (Q3): The third quartile is the value, where
75% of the values are smaller than Q3 and 25% are larger.

Outlier: usually, a value higher/lower than 1.5 x IQR

Descriptive Data Summarization

 Five Number Summary

Five number summary: min, Q1, Median, Q3, max

– Contain information about the endpoints (e.g., tails) of the
data

Boxplot
– Data is represented with a box
– The ends of the box are at the first and third quartiles, i.e., the height of
the box is IRQ
– The median is marked by a line within the box
– Whiskers: two lines outside the box extend to Minimum and Maximum
– To show outliers, the whiskers are extended to the extreme low and
high observations only if these values are less than 1.5 * IQR beyond
the quartiles.

Descriptive Data Summarization

 Five Number Summary

Boxplot for the unit price data for items sold at four branches of
AllElectronics during a given time period.

Descriptive Data Summarization

 Variance and Standard Deviation

Variance: (sample: s, population: σ)

1 n 1 n 2 1 n 2
2
s = ∑
n −1 i=1
2
( xi − x) = [∑ xi − (∑ xi ) ]
n −1 i=1 n i=1
n n
1 1 2
2
σ =
N
∑
i =1
2
( xi − µ ) =
N
∑ i
x
i =1
− µ 2

Standard deviation: s (or σ) is the square root of

variance s2 (or σ2)
– σ measures spread about the mean and should be used only
when the mean is chosen as the measure of center.
– σ =0 only when there is no spread, that is, when all
observations have the same value.
Descriptive Data Summarization
 Graphic Displays of Basic Descriptive Data
Summaries

Descriptive Data Summarization

 Graphic Displays

There are many types of graphs for the display of data

summaries and distributions, such as:
– Bar charts
– Pie charts
– Line graphs
– Boxplot
– Histograms
– Quantile plots
– Scatter plots
– Loess curves

Descriptive Data Summarization

 Histogram Analysis

Histograms or frequency histograms

– A univariate graphical method
– Consists of a set of rectangles that reflect the counts or
frequencies of the classes present in the given data
– If the attribute is categorical, such as automobile _model,
then one rectangle is drawn for each known value of A, and
the resulting graph is more commonly referred to as a bar
chart.
– If the attribute is numeric, the term histogram is preferred.

Descriptive Data Summarization

 Histogram Analysis

A set of unit price data for items sold at a branch of

AllElectronics.

Descriptive Data Summarization

 Quantile Plot

A quantile plot is a simple and effective way to have a

first look at a univariate data distribution.

Displays all of the data (allowing the user to assess
both the overall behavior and unusual occurrences)

Plots quantile information
– For a data xi data sorted in increasing order, fi indicates that
approximately 100 fi% of the data are below or equal to the
value xi

Note that the 0.25 quantile corresponds to quartile Q1,
the 0.50 quantile is the median, and the 0.75 quantile is
Q3.

Descriptive Data Summarization

 Quantile Plot

A quantile plot for the unit price data of AllElectronics.

Descriptive Data Summarization

 Scatter plot

A scatter plot is one of the most effective graphical

methods for determining if there appears to be a
relationship, clusters of points, or outliers between
two numerical attributes.

Each pair of values is treated as a pair of coordinates
and plotted as points in the plane

Descriptive Data Summarization

 Scatter plot

A scatter plot for the data set of AllElectronics.

Descriptive Data Summarization

 Scatter plot

Scatter plots can be used to find (a) positive or (b)

negative correlations between attributes.

Descriptive Data Summarization

 Scatter plot

Three cases where there is no observed correlation between the

two plotted attributes in each of the data sets.

Descriptive Data Summarization

 Loess Curve

Adds a smooth curve to a scatter plot in order to

provide better perception of the pattern of dependence

The word loess is short for local regression.

Loess curve is fitted by setting two parameters: a
smoothing parameter, and the degree of the
polynomials that are fitted by the regression

Descriptive Data Summarization

 Loess Curve

A loess curve for the data set of AllElectronics

Descriptive Data Summarization

 References

Descriptive Data Summarization

 References

J. Han, M. Kamber, Data Mining: Concepts and

Techniques, Elsevier Inc. (2006). (Chapter 2)

Descriptive Data Summarization

 The end

Descriptive Data Summarization