There is always a complex relationship between expert pages, while these relationships is the foundation of expert name disambiguation, traditional graph clustering method only consider the simple binary relation between expert pages, but this method always ignore multi relation between pages which are more complex, In order to utilize the relationship between expert pages efficiently, a Chinese expert name disambiguation approach based on hypergraph partitioning is proposed. Firstly, extract the characteristic attribute of expert in expert page. Secondly, construct a similarity matrix between the documents on different expert pages with the utilization of the attributes features and the associated relationship of the expert pages. Finally, set two kinds of constraints which are strong connection and negative connection, construct an expert page hypergraph model and use the method based on hypergraph partitioning to achieve expert name disambiguation. Through the contrast experiment in the Chinese expert disambiguation, it turns out that the disambiguation effect is much better with the adoption of hypergraph partitioning method.