Performance Analysis of Machine Learning and Deep Learning Models for Text Classification

Text classification is the task of forming semantic groups of text documents by assigning predefined class labels. It has wide range of real-life applications in various domains such as engineering, medical science, life science, social sciences and humanities, marketing, governance. Currently, machine learning and deep learning algorithms became popular and effective methods to address text classification problems with labelled data. In this work, we analyse the performance of different machine learning and deep learning algorithms for text classification. For this purpose, we selected six machine learning algorithms using three different vectorization techniques and five deep learning algorithms for the performance evaluation which is evaluated based on classification accuracy. All experiments are conducted on the 20 newsgroups dataset. Results indicate that Logistic Regression outperforms over other ML algorithms and a Bi-channel Convolution Neural Network model gains exciting results compared to other deep learning models.