Deep learning techniques for text classification

Abstract

This dissertation presents a series of experiments in applying deep learning techniques for text classification. The experiment will evaluate the performance of some popular deep learning models, such as feedforward, recurrent, convolutional, and ensemble-based neural networks, on five different datasets. We will build each model on top of two separate feature extractions to capture information within the text. The result shows that the word embedding provides a robust feature extractor to all the models in making a better final prediction. The experiment also highlights the effectiveness of the ensemble-based and temporal convolutional neural network in achieving good performances and even competing with the state-of-the-art benchmark models.