Recently, graph-based convolutional neural networks (GCNs) have attracted much attention by generalizing convolutional neural networks to graph data, which includes redefining the convolution and the pooling operations on graphs. Due to the limitation that graph data can only focus on dyadic relations, it cannot perform well in real practice. In contrast, a hypergraph can capture high-order data interaction and is easy to deal with complex data representation using its flexible hyperedges. Nevertheless, the existing methods for hypergraph convolutional networks are still not mature, currently there is no effective operation for hypergraph pooling. Therefore, a hypergraph pooling network with self-attention mechanism is proposed. Using hypergraph structure for data modeling, this model can learn hidden node features with high-order data information through hyper-convolution operation which introduces self-attention mechanism, select important nodes both on structure and content through hyper-pooling operation, and then obtain more accurate hypergraph representation. Experiments on text classification, dish classification, and protein classification tasks show that the proposed method outperforms recent state-of-the art methods.