The classical unsupervised clustering algorithm is fast, simple and suitable for mining large-scale datasets, and it can also directly divide communities. However, due to the complexity of communities, the classification accuracy of the algorithm is not ideal. Therefore, this paper proposed an error-correcting semi-supervised community detection algorithm(ESCD) based on active learning. It can calculate the traditional K-means algorithm step by step, and adding pairs of constraints in the clustering process. In order to preserve the correct partitioning according to the prior information, it corrected the wrong division to change the connection of the network. So that the network has a more obvious block structure in the process of changing the distance between nodes and cluster centers. The results of the experiment show that compared with the existing community discovery algorithms, the ESCD algorithm has higher accuracy with less supervisory information than other semi-supervised algorithms.

[2019-09-05] Printed Article /doc/2018.03.0162/v2