# Sequential Short-Text Classification with Recurrent and Convolutional Neural Networks

**Authors:** Ji Young Lee, Franck Dernoncourt  
**arXiv:** [1603.03827](https://arxiv.org/abs/1603.03827)  
**Date:** 2016-03-12  
**Summary page:** https://www.summarizepaper.com/en/paper/1603.03827v1/

## Key points

- Short-text classification is important in natural language processing
- Applications include sentiment analysis, question answering, and dialog management
- Artificial Neural Networks (ANNs) have shown promising results for short-text classification
- Existing ANN-based systems do not leverage preceding short texts when classifying a subsequent one
- Researchers from MIT developed a model based on Recurrent Neural Networks and Convolutional Neural Networks that incorporates the preceding short texts
- The proposed model achieves state-of-the-art results on three different datasets for dialog act prediction
- The study presents an important contribution to the field of Natural Language Processing by improving the accuracy of short-text classification through incorporating contextual information into ANNs.

## Summary

Short-text classification is a crucial task in natural language processing with applications in sentiment analysis, question answering and dialog management. Recent studies using Artificial Neural Networks (ANNs) have shown promising results for short-text classification. However, many short texts occur in sequences such as sentences in a document or utterances in a dialog and most existing ANN-based systems do not leverage the preceding short texts when classifying a subsequent one. To address this limitation, researchers from MIT developed a model based on Recurrent Neural Networks and Convolutional Neural Networks that incorporates the preceding short texts. The proposed model achieves state-of-the-art results on three different datasets for dialog act prediction and was accepted as a conference paper at NAACL 2016. This study presents an important contribution to the field of Natural Language Processing by improving the accuracy of short-text classification through incorporating contextual information into ANNs.