Chapter Objectives
In this chapter, you will learn to:
• understand the key steps of the analytics process model;
• identify the skill set of a data scientist;
• preprocess data for analytics using denormalization, sampling, exploratory data analysis, and dealing with missing values and outliers;
• build predictive analytical models using linear regression, logistic regression, and decision trees;
• evaluate predictive analytical models by splitting up the dataset and using various performance metrics;
• build descriptive analytical models using association rules, sequence rules, and clustering;
• understand the basic concepts of social network analytics;
• discern the key activities during post-processing of analytical models;
• identify the critical success factors of analytical models;
• understand the economic perspective on analytics by considering the total cost of ownership (TCO) and return on investment (ROI) and how they are affected by in- versus outsourcing, on-premise versus cloud solutions, and open versus commercial software;
• improve the ROI of analytics by exploring new sources of data, increasing data quality, securing management support, optimizing organizational aspects, and fostering cross-fertilization;
• understand the impact of privacy and security in a data storage, processing, and analytics context.
Opening Scenario
Now that Sober has made its first steps in business intelligence, it is eager to take this to the next level and explore what it could do with analytics. The company has witnessed extensive press and media coverage on predictive and descriptive analytics and wonders what these technologies entail and how they could be used to its advantage. It is actually thinking about analyzing its booking behavior, but is unsure how to tackle this. Given that Sober is a startup, it also wants to know the economic and privacy implications of leveraging these technologies.
In this chapter, we extensively zoom into analytics. We kick-off by providing a bird's eye overview of the analytics process model. We then give examples of analytics applications and discuss the data scientist job profile. We briefly zoom into data pre-processing. The next section elaborates on different types of analytics: predictive analytics, descriptive analytics, and social network analytics. We also discuss the post-processing of analytical models. Various critical success factors for analytical models are clarified in the following section. This is followed by a discussion on the economic perspective of analytics.
Review the options below to login to check your access.
Log in with your Cambridge Higher Education account to check access.
If you believe you should have access to this content, please contact your institutional librarian or consult our FAQ page for further information about accessing our content.