Question 1:
What is the Virtuous Cycle of Data Mining? describe in detail each of its four stages.
a. Identify the business problem: Finding the problem through the data. Talking to people
b. Transform data into actionable results: Using data to find business-oriented results.
c. Act on the knowledge: Including Insights, fixing data, real-time scoring etc….
d. Measure the results : Did data work and would output expect results in relation to the business?
Question 2:
• What is a good business goal? Break the business goal down in terms of data mining tasks, and explain one of the tasks for that specific goal.
• List and explain what the six phases of The Cross-Industry Standard for Data Mining do and how might they affect a business ?
Question 3:
(1.) Identify and explain the:
(a.) Advantages of decision trees
(b.) Applications of decision trees
(2.) (a) Give two examples of decision tree algorithms.
(b.) Explain the challenges that might be faced using these algorithms.
(c.) Discuss potential applications of decision trees to business data mining.
Question 4:
- List all five steps to applying neural networks to applications and explain and discuss was it entailed in each step.
- How does a neural network learn with back propagation? List the three ways and discuss and expand on each step.
- Where can neural networks be used in practice and discuss how do they work in that industry?
Question 5:
(a). What purpose does cluster analysis serve, and discuss how does it benefit
analysts?
(b.) What are the most widely used clustering methods? Explain what they do.
(c.) What are the variations on K-Means? Explain the uses.
(d.) What are two real applications of cluster analysis? Explain how cluster
analysis is used in these applications.
