ROC CURVE ANALYSIS OF DIFFERENT HYBRID FEATURE DESCRIPTORS USING MULTI CLASSIFIERS

Tremendous success of machine learning algorithms at pattern recognition creates interest in new inventions. Machine an era big data is that significant hierarchical relationships within the can be discovered algorithmically than other handcraft like features. In this study, Convolutional Neural Network (CNN) used as feature descriptors pulmonary malignancy prediction. Various such Histogram Oriented Gradient (HOG), Extended (EXHOG) and Linear Binary Pattern (LBP) are analyzed with classifiers Random Forest (RF), Decision Tree (DT), K-Nearest Neighbour (KNN) Support Vector (SVM) for Computed Tomography (CT) The phenotype features nodules important cues identification. nodule solidity cue white blob area method Lung Image Database Consortium (LIDC) dataset. Receivers Operating Characteristics (ROC) curves show graphical summaries detectors performance. It proved CNN based extraction SVM classifier works well