Semi-supervised clustering takes advantage of a small amount of labeled data to bring a great benefit to the clustering of unlabeled data. Based on a locality sensitive k-means clustering method, this paper presents two novel semi-supervised clustering algorithms inspired by the semi-supervised variants of the k-means clustering by seeding. To investigate the effectiveness of our approaches, experiments are done on one artificial dataset and three real datasets. Experimental results show that two proposed methods can improve the clustering performance significantly compared to other unsupervised and semi-supervised clustering algorithms.
展开▼