Data Analytics: Exploiting the Data Warehouse - Data Warehouse Systems: Design and Implementation - page 373

rules at different levels of granularity. The pattern growth algorithm was
introduced in [ 76 ]. Sequential pattern algorithms are described in [ 4 , 191 ].
Variants to make them more ecient by pruning uninteresting sequences are
explained in Garofalakis et al. [ 59 , 60 ]. Clustering is studied in [ 101 ], and
the DBSCAN algorithm is described in [ 51 ]. The classic SLIQ and SPRINT
classification algorithms are described in [ 131 , 186 ], respectively.
Most of the topics on KPIs and dashboards are oriented to practitioners.
A typical reference on KPIs is the topic by Parmenter [ 156 ]. Some references
for dashboards are [ 55 , 56 , 162 , 169 ]. Details on Microsoft Reporting Services
can be found in [ 209 ].
9.6 Review Questions
9.1 What is data mining? Which disciplines does it comprise? How does
data mining differentiate from statistics?
9.2 How does data mining fit into the process of knowledge discovery in
databases?
9.3 Describe the main data mining tasks.
9.4 Describe the components of most data mining algorithms.
9.5 What is the main difference between supervised and unsupervised
classification? Give examples of the possible uses of each technique.
9.6 What are decision trees? What is the Gini index used for?
9.7 Explain the partition algorithm for building a decision tree.
9.8 What is clustering? How would you select a good distance function?
What is a score function?
9.9 Explain the K-means algorithm for clustering.
9.10 What is an association rule? Define confidence and support. What are
they used for? What is an itemset? What is a frequent itemset?
9.11 Explain the Apriori algorithm for mining association rules.
9.12 Explain the concept of hierarchical association rules. How would you
use them in a data mining process?
9.13 Describe the FUP algorithm for updating association rules. Give the
principles on which this algorithm works and discuss its limitations.
9.14 Explain the concept of mining sequential patterns. How does it
differentiate from mining association rules? What are the implications
of accounting for order?
9.15 What are key performance indicators or KPIs? What are they used for?
Detail the conditions a good KPI must satisfy.
9.16 Define a collection of KPIs using an example of an application domain
that you are familiar with.
9.17 Explain the notion of dashboard. Compare the different definitions for
dashboards.
9.18 What types of dashboards do you know? How would you use each kind?