Data Mining and Predictive Analytics, 2nd Edition

Appendix A

Data Summarization and Visualization

Here we present a very brief review of methods for summarizing and visualizing data. For deeper coverage, see Discovering Statistics, Second Edition, by Daniel Larose (W.H. Freeman, second edition, 2013).

Part 1: Summarization 1: Building Blocks Of Data Analysis

• Descriptive statistics refers to methods for summarizing and organizing the information in a data set.

  Consider Table A.1, which we will use to illustrate some statistical concepts.

  Table A.1 Characteristics of 10 loan applicants

  Applicant	Marital Status	Mortgage	Income ($)	Rank	Year	Risk
  1	Single	Y	38,000	2	2009	Good
  2	Married	Y	32,000	7	2010	Good
  3	Other	N	25,000	9	2011	Good
  4	Other	N	36,000	3	2009	Good
  5	Other	Y	33,000	4	2010	Good
  6	Other	N	24,000	10	2008	Bad
  7	Married	Y	25,100	8	2010	Good
  8	Married	Y	48,000	1	2007	Good
  9	Married	Y	32,100	6	2009	Bad
  10	Married	Y	32,200	5	2010	Good

• The entities for which information is collected are called the elements. In Table A.1, the elements are the 10 applicants. Elements are also called cases or subjects.
• A variable is a characteristic of an element, which takes on different values for different elements. The variables in Table A.1 are marital status, mortgage, income, rank, year, and risk. Variables are also called attributes.
• The set of variable values for a particular element is an observation. Observations are also called records. The observation for Applicant 2 is:

Applicant ...