In plain words: A text classifier reads documents letter by letter, using a sliding window to spot patterns, so it never has to split words apart. On several large datasets it matched or beat the usual word-counting methods and networks that read whole words.

Abstract · Character-level Convolutional Networks for Text Classification

This article offers an empirical exploration on the use of character-level convolutional networks (ConvNets) for text classification. We constructed several large-scale datasets to show that character-level convolutional networks could achieve state-of-the-art or competitive results. Comparisons are offered against traditional models such as bag of words, n-grams and their TFIDF variants, and deep learning models such as word-based ConvNets and recurrent neural networks.

Xiang Zhang, Junbo Zhao, Yann LeCun
arXiv:1509.01626 · cs.LG, cs.CL · submitted Sep 4, 2015 · updated Apr 4, 2016
abstract · pdf · html · An early version of this work entitled "Text Understanding from Scratch" was posted in Feb 2015 as arXiv:1502.01710. The present paper has considerably more experimental results and a rewritten introduction, Advances in Neural Information Processing Systems 28 (NIPS 2015)