Better Decision Tree from Intelligent Instance Selection

This book describes theoretical and experimental
studies of instance selection to improve data mining
model. Data preparation is one of the most important
and time consuming phases in knowledge discovery.
Preparation tasks often determine the success of data
mining engagements. The importance of instance
selection is the primary focus because the size of
current and future databases often exceeds the amount
of data which current data mining algorithms can
handle properly. Instance selection thus can be used
to improve scalability of data mining algorithms as
well as improve the quality of the data mining results.

This book presents a new optimization-based approach
for instance selection that uses a genetic algorithm
to select a subset of instances to produce a simpler
decision tree model with acceptable accuracy. The
resultant trees are easier to comprehend and
interpret by the decision maker and hence more useful
in practice. Numerical results are obtained for
several difficult test data sets indicating that
GA-based instance selection can often reduce the size
of the decision tree by an order of magnitude while
still maintaining good prediction accuracy.

Autorentext

Dr. Wu has over 10 years experiences in data management, data mining, and operations research. His expertise is in instance selection, decision tree modeling and metaheuristic optimization. He holds a bachelor's and master's degree from Tsinghua University in China, and a PhD's degree in Industrial Engineering from Iowa State University.

Klappentext

This book describes theoretical and experimentalstudies of instance selection to improve data miningmodel. Data preparation is one of the most importantand time consuming phases in knowledge discovery.Preparation tasks often determine the success of datamining engagements. The importance of instanceselection is the primary focus because the size ofcurrent and future databases often exceeds the amountof data which current data mining algorithms canhandle properly. Instance selection thus can be usedto improve scalability of data mining algorithms aswell as improve the quality of the data mining results. This book presents a new optimization-based approachfor instance selection that uses a genetic algorithmto select a subset of instances to produce a simplerdecision tree model with acceptable accuracy. Theresultant trees are easier to comprehend andinterpret by the decision maker and hence more usefulin practice. Numerical results are obtained forseveral difficult test data sets indicating thatGA-based instance selection can often reduce the sizeof the decision tree by an order of magnitude whilestill maintaining good prediction accuracy.