Xiang Zhang, Junbo Zhao, Yann Lecun
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.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Sentiment Analysis | Yelp Fine-grained classification | Error | 37.95 | Char-level CNN |
| Sentiment Analysis | Yelp Binary classification | Error | 4.88 | Char-level CNN |
| Text Classification | DBpedia | Error | 1.55 | Char-level CNN |
| Text Classification | AG News | Error | 9.51 | Char-level CNN |
| Classification | DBpedia | Error | 1.55 | Char-level CNN |
| Classification | AG News | Error | 9.51 | Char-level CNN |