Data Classification using Modified Version of Support Vector Machine

Computer Engineering

Text classification is a process of categorized data according to their class. With the instant growth of information, text classification has become the vital techniques for handling and organizing text data. In general, text classification plays an important role in information extraction and summarization, text retrieval, and question-answering such as medical diagnosis, news group filtering, spam filtering, and sentiment analysis. The process of classification consists four stages: text preprocessing, feature extraction, training classifier and training model. In this first stage the dataset is divided into training data and testing data. After that data is preprocessed and features are extracted from that data then after classification model is constructed. The data is classified using machine learning techniques and statistical techniques such as k-nearest neighbors, support vector machine, naive Bayesian method. The classification task also exemplifies the hybrid approach of text classification techniques. In this research we provide a modified version of support vector machine with better membership function for text classification of text data. In that the noisy and inherent data is handled by fuzzy support vector machine and its fuzzy membership function which is used hyperbolic tangent kernel.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.