Image-based Natural Language Understanding Using 2D Convolutional Neural Networks

TL;DR

This paper introduces a novel method for natural language understanding by transforming text into images and applying 2D CNNs, eliminating the need for OCR and sequential models, and demonstrating superior performance in classification and dialog tasks.

Contribution

The paper presents a new image-based approach to NLP that leverages 2D CNNs to learn semantic features directly from text images, outperforming traditional methods.

Findings

Outperformed state-of-the-art in non-Latin text classification

Achieved promising results on multiple datasets

Surpassed memory networks on out-of-vocabulary dialog entities

Abstract

We propose a new approach to natural language understanding in which we consider the input text as an image and apply 2D Convolutional Neural Networks to learn the local and global semantics of the sentences from the variations ofthe visual patterns of words. Our approach demonstrates that it is possible to get semantically meaningful features from images with text without using optical character recognition and sequential processing pipelines, techniques that traditional Natural Language Understanding algorithms require. To validate our approach, we present results for two applications: text classification and dialog modeling. Using a 2D Convolutional Neural Network, we were able to outperform the state-of-art accuracy results of non-Latin alphabet-based text classification and achieved promising results for eight text classification datasets. Furthermore, our approach outperformed the memory networks when using out of vocabulary entities fromtask 4 of the bAbI dialog dataset.