Home | Search | Help  
Home Page Università di Genova

Technical Report Details

Date 12-6-2006
Number DISI-TR-06-10
Title A wrapper approach to supervised input selection using simulated annealing
Authors Maurizio Filippone, Francesco Masulli, Stefano Rovetta
Bibtex Entry @techreport{FilipponeTR06, author = "Maurizio Filippone and Francesco Masulli
E-mail filippone@disi.unige.it
Link ftp://ftp.disi.unige.it/person/FilipponeM/Publications/tech_rep_ijar06.pdf
Abstract Genomic data, and more generally biomedical data, are usually characterized by high dimensionality. A feature selection procedure can attain the two objectives of highlighting the relevant variables (genes) and improving classification results. In this paper we propose a wrapper approach to gene selection in classification of gene expression data using Simulated Annealing coupled with supervised classifiers. The proposed approach can perform global combinatorial searches through the space of all possible input subsets, can handle cases with numerical, categorical or mixed inputs, and is able to find (sub-)optimal subsets of inputs giving very low classification errors. The method has been tested on publicly available bioinformatics data sets, and also on mixed type data, using Support Vector Machines or Classification Trees. Moreover we propose some heuristics able to speed-up the convergence. The experimental results highlight the ability of the method to select minimal sets of relevant genes.
Back to Technical Reports