A Comparison of Neural Network Training Methods for Text Classification

TL;DR

This paper evaluates neural network training techniques for Twitter weather classification, demonstrating that deep neural networks outperform SVMs with feature extraction and various training optimizations.

Contribution

It introduces a comprehensive comparison of training methods for deep neural networks in text classification, highlighting the benefits of layer-wise pretraining and advanced optimization.

Findings

Deep neural networks outperform SVMs with Gaussian kernels.

Adding hidden layers improves performance.

Nesterov's Accelerated Gradient enhances training efficiency.

Abstract

We study the impact of neural networks in text classification. Our focus is on training deep neural networks with proper weight initialization and greedy layer-wise pretraining. Results are compared with 1-layer neural networks and Support Vector Machines. We work with a dataset of labeled messages from the Twitter microblogging service and aim to predict weather conditions. A feature extraction procedure specific for the task is proposed, which applies dimensionality reduction using Latent Semantic Analysis. Our results show that neural networks outperform Support Vector Machines with Gaussian kernels, noticing performance gains from introducing additional hidden layers with nonlinearities. The impact of using Nesterov's Accelerated Gradient in backpropagation is also studied. We conclude that deep neural networks are a reasonable approach for text classification and propose further ideas to improve performance.