A Novel Method for DNA Microarray Data Analysis: SDL Global Optimization Method - DNA Microarray Technology and Data Analysis in Cancer Research - page 37

CHAPTER 4
A Novel Method for DNA Microarray
Data Analysis: SDL Global
Optimization Method
This chapter is concerned with the challenge of mining knowledge from
DNA microarray gene expression data. With the objective to discover
unknown patterns from microarray data, methodologies are derived from
machine learning, artificial intelligence, and statistics. Nowadays,
microarray expression data accumulate at an alarming speed in various
storage devices, and so does valuable information. However, it is difficult
to understand information hidden in data without the aid of data analysis
techniques. Both machine learning and data mining have been applied to
the field in order to better understand microarray expression datasets.
A data mining system usually enables one to collect, store, access, process,
and ultimately describe and visualize datasets. The discussion of data
collection and storage is not included here, though it is important for
mining microarray expression data.
Data mining has successfully provided solutions for finding informa-
tion from data in many medical research fields, such as bioinformatics and
pharmaceuticals. Many important problems have been addressed by data
mining methods, such as neural networks, fuzzy logic, decision trees,
genetic algorithms, and statistical methods. Data mining tasks can be
descriptive and predictive; in other words, it is an interdisciplinary field
with a general goal of predicting outcomes and uncovering relationships
in data (Han and Kamber, 2001; Hand et al ., 2001; Kantardzic, 2002;
Mitra et al ., 2002). Microarray data analysis is one of the most attractive
fields of data mining. With the help of gene expressions obtained from
37