ABSTRACT
Genetic Programming is a method to automatically create computer programs based on the principles of evolution. The problem of deceptiveness caused by complex dependencies among components of programs is challenging. It is important because it can misguide Genetic Programming to create suboptimal programs. Besides, a minor modification in the programs may lead to a notable change in the program behaviours and affect the final outputs. This article presents Grammar-Based Genetic Programming with Bayesian Classifiers (GBGPBC) in which the probabilistic dependencies among components of programs are captured using a set of Bayesian network classifiers. Our system was evaluated using a set of benchmark problems (the deceptive maximum problems, the royal tree problems, and the bipolar asymmetric royal tree problems). It was shown to be often more robust and more efficient in searching the best programs than other related Genetic Programming approaches in terms of the total number of fitness evaluation. We studied what factors affect the performance of GBGPBC and discovered that robust variants of GBGPBC were consistently weakly correlated with some complexity measures. Furthermore, our approach has been applied to learn a ranking program on a set of customers in direct marketing. Our suggested solutions help companies to earn significantly more when compared with other solutions produced by several well-known machine learning algorithms, such as neural networks, logistic regression, and Bayesian networks.
Subject(s)
Algorithms , Neural Networks, Computer , Bayes Theorem , Machine Learning , SoftwareABSTRACT
The global increase of gene expression has been frequently established in cancer microarray studies. However, many genes may not deliver informative signals for a given experiment, due to insufficient expression or even non-expression, despite the DNA microarrays massively measuring genes in parallel. Hence the informative gene set, rather than the whole genome, should be more reasonable to represent the genome expression level. We observed that the trend of over-expression for informative genes is more obvious in human cancers, which is to some extent masked using the whole genome without any filtering. Accordingly we proposed a novel normalization method, Informative CrossNorm (ICN), which performs the cross normalization (CrossNorm) on the expression matrix merely containing the informative genes. ICN outperforms other methods with a consistently high precision, F-score, and Matthews correlation coefficient as well as an acceptable recall based on three available spiked-in datasets with ground truth. In addition, nine potential therapeutic target genes for esophageal squamous cell carcinoma (ESCC) were identified using ICN integrated with a protein-protein interaction network, which biologically demonstrates that ICN shows superior performance. Consequently, it is expected that ICN could be applied routinely in cancer microarray studies.
