A novel inductive ensemble clustering method is proposed. In the proposed method, kernel support matching is applied to a co-association matrix that aggregates arbitrary basic partitions in order to detect clusters of complicated shape. It also has the advantage of naturally detecting the number of clusters and assigning clusters for out-of-sample data. In the proposed method, a new similarity is learned from various clustering results of the basic partition, and a kernel support function capable of clustering learning data and test data is constructed. Experimental results demonstrated that the proposed method is effective with respect to clustering quality and has the robustness to induce clusters of out-of-sample data.