A hybrid heuristic dimensionality reduction technique for microarray gene expression data classification: a blending of GA, PSO and ACO. (2016)

• Record Type: Journal Article
• Authors: Uma, S.M.; Kirubakaran, E.
• Is Part Of: International journal of data mining, modelling and management. Volume 8:Number 2(2016)
• Page Start: 160
• Page End: 179
• Publication Date: 2016
• Subjects: microarrays -- gene expression data -- principal component analysis -- PCA -- genetic algorithms -- GAs -- particle swarm optimisation -- PSO -- ant colony optimisation -- ACO -- dimensionality reduction -- data classification -- bioinformatics -- acute lymphoblastic leukaemia -- ALL -- acute myeloid leukaemia -- AML; Data mining -- Periodicals; Information science -- Periodicals; Databases -- Periodicals; 005.7
• Languages: English
• ISSNs: 1759-1163

Abstract

The micro-array gene expression data are mostly utilised for biological applications for classification of tumours in genes. Majority of recent work are concentrated to reduce the dimension of micro-array gene expression data. The modern dimension reduction approaches have followed statistical techniques and transformation techniques like principal component analysis (PCA). This paper deals with a hybrid heuristic dimensionality reduction technique for microarray gene expression data classification using biologically inspired heuristic algorithms. The proposed hybrid heuristic technique has combined genetic algorithm (GA), particle swarm optimisation (PSO) and ant colony optimisation (ACO). The dimensionality reduced data using the proposed technique correlates well with the classifier with less classification error after training. The test case investigation is performed with acute lymphoblastic leukaemia (ALL) and acute myeloid leukaemia (AML) dataset and the classification accuracy of the proposed technique are compared with conventional technique.