Comparison of Machine Learning Algorithms for Identifying Cancer Types Side: 1 of 1
The following text was automatically extracted from the image on this page using optical character recognition software:
Microarray technologies helps in visualization of
expression of thousands of genes at a glance. This is
exceedingly helpful in studying a disease like cancer
where interplay of various genes results in varied
types of tumors. This technology generates large
amounts of data that is difficult to effectively analyze
manually. Machine learning algorithms, such as
random forests, have been shown to effectively
predict useful genes and types of cancer from
In this study, we propose using additional machine
learning algorithms, such as artificial neural
networks, support vector machines, and random
forests to analyze gene expression datasets to
quickly and accurately identify types of cancers. We
shall compare how these additional methods
compare to random forests as proposed in the
previous study(1). We will also be testing how
changing the parameters for these algorithms affects
0 Machine Learning Algorithms
0 Artificial Neural Networks (ANN)
0 Computer models designed to imitate the human
brain for decision making tasks. The ANN using
various learning rates, .1, .5, .9, and numbers of
hidden layers, 0, 1, 2, with 0, 20, and 15 nodes
per hidden layer respectively. Additionally, the
momentum value was kept at 1 to test the
learning rates in isolation.
0 Random Forests (RF)
0 An ensemble learning method for classifying
data. It constructs a series of randomly generated
trees during training and the most frequent
output class is considered the correct
classification. In this study, 14 different random
genes were selected for each run, and the
number of trees was alternated between 10, 50,
100, 150, 200, 250, 300 for each dataset.
flat; I i ?'t'ivrv' spf.icr1
i " "
[Crnnun Lsi et a1. 2011]
Figure 3: Basic Working of Random Forests
" The data sets we used were courtesy of
(1). They represent multiple types of
cancer each with various classes.
" Breast2 Cancer Growth and Progression
Colon Normal Cell
orm ,Co, Mutation results
0 Normal Cell Growth in Cancer
' Cancerous Growth forms Tumo
NCI I r
" Prostate Tumors bud and spread throughout the body
" S R B C T http://www.unc.edu/depts/our/hhmi/hhmi-ft-learningmodules/cancermodule/images/cancergrowth.
Figure 4: Cancer Growth and its Proliferation
Graph 2: Graph
Graph 3: Graph
70.00% * I
0 Each of the instances for each dataset were
classified by the machine learning algorithms
4:.. L0 . La 0 I If0
This poster can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this side or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current side of this Poster.
Saxena, Garima; Helsing, Joseph; Reyes, Omar Costilla & Azad, Rajeev K. Comparison of Machine Learning Algorithms for Identifying Cancer Types, poster, March 2014; (digital.library.unt.edu/ark:/67531/metadc330322/m1/1/: accessed November 17, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT College of Engineering.