Neural Network Ensemble with Negatively Correlated Features for Cancer Classification

Abstract

Abstract. The development of microarray technology has supplied a large volume of data to many fields. In particular, it has been applied to prediction and diagnosis of cancer, so that it expectedly helps us to exactly predict and diagnose cancer. It is essential to efficiently analyze DNA microarray data because the amount of DNA microarray data is usually very large. Since accurate classification of cancer is very important issue for treatment of cancer, it is desirable to make a decision by combining the results of various expert classifiers rather than by depending on the result of only one classifier. In spite of many advantages of ensemble classifiers, ensemble with mutually errorcorrelated classifiers has a limit in the performance. In this paper, we propose the ensemble of neural network classifiers learned from negatively correlated features to classify cancer precisely, and systematically evaluate the performance of the proposed method using three benchmark datasets. Experimental results show that the neural network ensemble with negatively correlated features produces the best recognition rate on the three benchmark datasets.

Indexing ICD-9 codes for free-textual clinical diagnosis records by a new ensemble classifier

“…Determining learning algorithms for the sake of producing classifiers and using an ensemble fusion method are the principal tasks in an ensemble classifier (Kuncheva, 2002). Other examples of ensemble classifier can be found in Banfield et al (2007), derWalt and Barnard (2006), Hansen and Salamon (1990), Plewczynski et al (2006), Wang et al (2003) and Won and Cho (2003).…”

Statistical data processing in clinical proteomics

“…A combination of bagging and boosting is used by Dettling in BagBoosting, where in each boosting step a bagged classifier is constructed [68]. Alternatively, resampling of the variables also leads to diverse base classifiers [69][70][71][72]. After construction of the base classifiers, their diversity can be evaluated by comparing their predictions [60,69] or the structure of the individual classifiers [63].…”

Lymphoma Cancer Classification Using Genetic Programming with SNR Features

“…Since not all the genes are associated with a specific disease, the feature selection often called gene selection is necessary to extract informative genes for the classification of the disease [3,15,16]. Moreover, feature selection accelerates the speed of learning a classifier and removes noises in the data.…”

“…It is simultaneously done by the training of a classifier to produce the optimal combination of features and a classifier. Since the filter approach is simple and fast enough to obtain high performance, we evaluated various filter-based feature selection methods [15]. Finally signal-to-noise ratio ranking method is adopted to select useful features.…”