Exploration on Document Taxonomy by Ganb Algorithm

Abstract

In this research, we propose an integrated classification GANB algorithm that combines a feature extractor with a classifier to construct a classification model. The feature extractor automates the examination of raw pre-processed unstructured documents. Following feature extraction, categorization generates meaningful classes based on the supplied features. The study uses a genetic algorithm (GA) for feature extraction and Naïve Bayes(NB) for classification purposes. The simulation evaluates the suggested classification model's accuracy, sensitivity, specificity, and f-measure using various performance indicators. Over the Medline cancer datasets, the suggested GANB gets a higher classification rate than existing approaches.