Named entity recognition for Chinese microblog with convolutional neural network

Abstract

Named Entity Recognition (NER) has usually focused on traditional formal text. we consider the task of NER on microblog text. In this paper, we propose a Convolutional Neural Network for NER in Chinese microblog text. Instead of traditional machine learning needing man-made input features carefully optimized for NER task, our system learns the words feature by itself. Our network uses a sliding window of word context to predict tags. Experimental results show that our model achieved 80% accuracy on this task.