ESBO: CANCER CLASSIFICATION IN MICROARRAY DATA USING SBO ALGORITHM WITH SELF-ADAPTIVE EMISSION RATE
Keywords:
Classification, Stellar Mass Black hole, ene Expression data,, Emission Rate, Optimization.Abstract
The complete diagnosis of sub type of cancers underpins the care of individual cancer patients. In the perspective of cancer classification, gene
expression data analysis has been used to more precisely classify tumors. However, selection of non-redundant but relevant genes from microarray gene
expression data is computationally tough task. It has become increasingly clear that the traditional approach to cancer classification is insufficient. The
major motto of this article is to originate a deterministic approach to pick the highly relevant genes from the microarray data for cancer diagnosis. This
article presents a modified nature-inspired algorithm namely Stellar-Mass Black hole with Self-Adaptive Emission Rate (ESBO) to choose genes from
microarray data that are able to classify various cancer sub-types with high accuracy. The experiment done and results are analyzed with five reputed
Micro array benchmark datasets. The results depict that ESBO is outshines than SBO and other deterministic methods. The experiment also proves that
adopting dynamic adjustment of emission rate with SBO is more effective than applying them individually.
