# Efficient Character-level Document Classification by Combining Convolution and Recurrent Layers

arXiv: https://arxiv.org/abs/1602.00367 · HTML: https://mlsanity.com/paper/1602.00367/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Feb 2016.

## Built on by (most cited first, 6 of 6)

- [Hierarchical Neural Network Approaches for Long Document Classification](https://mlsanity.com/paper/2201.06774/) (2022)
- [Auxiliary Cross-Modal Representation Learning with Triplet Loss Functions for Online Handwriting Recognition](https://mlsanity.com/paper/2202.07901/) (2022)
- [Deep Learning-based Sentiment Classification: A Comparative Survey](https://mlsanity.com/paper/2312.17253/) (2023)
- [An Attention Ensemble Approach for Efficient Text Classification of Indian Languages](https://mlsanity.com/paper/2102.10275/) (2021)
- [Embedding Compression for Text Classification Using Dictionary Screening](https://mlsanity.com/paper/2211.12715/) (2022)
- [Unsupervised Threat Hunting using Continuous Bag-of-Terms-and-Time (CBoTT)](https://mlsanity.com/paper/2403.10327/) (2024)

## Abstract (arXiv)

Document classification tasks were primarily tackled at word level. Recent research that works with character-level inputs shows several benefits over word-level approaches such as natural incorporation of morphemes and better handling of rare words. We propose a neural network architecture that utilizes both convolution and recurrent layers to efficiently encode character inputs. We validate the proposed model on eight large scale document classification tasks and compare with character-level convolution-only models. It achieves comparable performances with much less parameters.