Business Intelligence and Information Systems for Decision Making Chapter 9 Business Intelligence and Information Systems for Decision Making
This Could Happen to You: “We’re Sitting on All This Data” Anne proposes to combine membership data and publicly available data in order to better target marketing efforts for Fox Lake weddings. Information will allow her classy promotions and increase wedding revenue. Scenario video Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Study Questions Q1: Why do organizations need business intelligence? Q2: How do business intelligence (BI) systems provide competitive advantages? Q3: What problems do operational data pose for BI systems? Q4: What are the purpose and components of a data warehouse? Q5: What is a data mart and how does it differ from a data warehouse? Q6: What are the characteristics of data-mining systems? How does the knowledge in this chapter help Fox Lake and you? Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q1: Why Do Organizations Need Business Intelligence? Businesses collect massive amounts of data Reveal important patterns of relationships and valuable information buried in that data Data communications and data storage essentially free 2 million emails, 31,000 text messages, and 162,000 instant messages transmitted every second (2007) 2010 total online computer storage about 600 exabytes 70 exabytes equivalent to 14 times total number of words ever spoken by humans Such as, evidence that someone is going to default on a loan Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
How Big Is an Exabyte? Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q2: How Do Business Intelligence Systems Provide Competitive Advantages? Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Business Intelligence Tools vs. Business Intelligence Systems Q2: How Do Business Intelligence Systems Provide Competitive Advantages? (cont’d) Business Intelligence Tools vs. Business Intelligence Systems BI Tools (software) Crystal Reports SPSS Clementine SharePoint Server BI Systems Reporting System Datamining System Knowledge Mgmt System Expert System Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q3: What Problems Do Operational Data Pose for BI Systems? Raw data usually unsuitable for sophisticated reporting or data mining Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Key Terms Data granularity—degree of summarization or detail. Coarse data highly summarized; fine data precise details Clickstream data—customers’ website clicking behavior Curse of dimensionality—too much data (attributes/columns or rows) Market-basket analysis—computes correlations of items on past orders to determine items frequently purchased together Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q4: What Are the Purpose and Components of a Data Warehouse? Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Consumer Data Available for Purchase from Data Vendors Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Sources of Data for Data Warehouses Internal operations systems External data purchased from outside sources Data from social networking, user-generated content applications Metadata concerning data stored in data warehouse Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q5: What Is a Data Mart and How Does It Differ from a Data Warehouse? Data Mart => Collection of data created to address needs of a particular: Business function Problem Opportunity Marts created from data extracted from data warehouse Data mart is like a retail store in a supply chain Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Components of a Data Mart Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Q6: What Are the Characteristics of Data-Mining Systems? Data mining—application of statistical techniques to find patterns and relationships in body of data for purpose of classifying and predicting Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Unsupervised Data Mining Analysts do not create model before running analysis Apply data-mining technique and observe results Hypotheses created after analysis as explanation for results Technique: Cluster analysis to find groups with similar characteristics Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Supervised Data Mining Model developed before analysis Statistical techniques used to estimate variable parameters Regression analysis—measures impact of set of variables on one another Example: CellPhoneWeekendMinutes = 12 X (17.5 X CustomerAge) + (23.7 X NumberMonthsOfAccount) Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Supervised Data Mining (cont’d) Neural networks Used for predicting values and making classifications Complicated set of nonlinear equations See www.kdnuggets.com to learn more Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall
Active Review Q1: Why do organizations need business intelligence? Q2: How do business intelligence (BI) systems provide competitive advantages? Q3: What problems do operational data pose for BI systems? Q4: What are the purpose and components of a data warehouse? Q5: What is a data mart and how does it differ from a data warehouse? Q6: What are the characteristics of data-mining systems? How does the knowledge in this chapter help Fox Lake and you? Copyright © 2012 Pearson Education, Inc. Publishing as Prentice Hall