Abstract: A new method for the measurement of entity relevance was proposed on the basis of entity relationship graph. By defining the area graph, the semantic of each entity was well presented. To speed up the calculation of the similarity between two area subgraphs, the area subgraphs were first converted into semantic trees. Then the kernel function was used to calculate the similarity by counting the number of shared subtrees. At last, the proposed methods were evaluated on the basis of the experimental results. The experimental results proved that the proposed method had a good performance in both accuracy and efficiency.