In this paper, we propose a question representation based on entity labeling and question classification for a automatic question answering system of Chinese Gaokao history question. A CRF model is used for the entity labeling and SVM/CNN/LSTM models are tested for question classification. Our experiments show that CRF model provides a high performance when used to label informative entities out while neural networks has a promising performance for the question classification task. With both entity labeling and question classification models of high performance, we can provide the KB-based question answering system with a question representation of high reliability. Then the question answering system can do more good work depending on the key information our models provide.