An efficient SVM based tumor classification with symmetry Non-negative Matrix Factorization using gene expression data

Abstract

A reliable and accurate identification of the type of tumors is crucial to the proper treatment of cancers. The classification of tumors was and is both a practical and theoretic necessity and requirement. DNA microarrays provide a new technique of measuring gene expression, which has attracted a lot of research interest in recent years. It was suggested that gene expression data from microarrays (biochips) can be employed in many biomedical areas, e.g., in cancer classification. Although several, new and existing, methods of classification were tested, a selection of proper (optimal) set of genes, the expressions of which can serve during classification, is still an open problem. This paper presents a new method for tumor classification using gene expression data. In the proposed method, we first select genes using Nonnegative Matrix Factorization (NMF). In order to improve the performance of classification, Symmetry NMF (SymNMF) is used in this approach. Then, features are extracted from the selected genes by virtue SymNMF. As a last step, an efficient machine learning approach is used to classify the tumor samples using the extracted features. In order for a better classification, Support Vector Machine with Weighted Kernel Width (WSVM) is used in this classification approach. The performance of the proposed approach is tested using colon cancer data set and the acute leukemia data set. It is observed from the experimental results that the proposed approach provides better performance when compared with the traditional approaches.

A survey of pattern classification-based methods for predicting survival time of lung cancer patients

“…In [15], Brunet et al used NMF to cluster samples and genes molecular pattern discovery. In [16], Yuvaraj et al used it to identify distinct molecular patterns of cancer for class discovery.…”

“…It has been proven that SVMs are good at dealing with the difficulty of "high dimensionality and small sample size" in practice [24,25]. Over decades, SVMs have been applied to a variety of application fields such as image processing [26], texture detection [27] and tumor classification [16,21,22,24,28,29]. In implementing, SVMs have two important parts to be pre-specified: kernel function and cost parameter.…”

“…SVMs, as a state-of-the-art method, were developed based on the principle of structural risk minimization [16]. Theoretically, SVMs attempt to construct an optimal hyperplane by mapping data into a high-dimensional space for classification and regression.…”

Malignant and Benign Brain Tumor Segmentation and Classification Using SVM with Weighted Kernel Width

“…For the purpose of diminishing the coexisting over-fitting and under-fitting loss in support vector classification using Gaussian RBF kernel, the kernel width is needed to be adjusted, to some extent, the feature space distribution. The scaling rule is that in dense regions the width will be narrowed (through some weights less than 1) and in sparse regions the width will be expanded (through some weights more than 1) [20]. The Weighted Gaussian RBF kernel is as follows:…”

Integrating Type-1 Fuzzy and Type-2 Fuzzy Clustering with K-Means for Pre-Processing Input Data in Classification Algorithms

“…Since the classes are determined before applying the real data, this method is known as a supervised learning algorithm. Classification is used in many fields and sciences such as, image segmentation [1,2], geology [3], robot control [4,5], bio-informatics [6], genetics [8], biology [7] and healthcare [9].…”