Semi-supervised k-means clustering for multi-type relational data

In many data mining tasks, there is a large supply of unlabeled data but limited labeled data since it is expensive generated. Therefore, a number of semi-supervised clustering algorithms have been proposed, but few of them are specially designed for multi-type relational data. In this paper, a semi-supervised k-means clustering algorithm for multi-type relational data is proposed, which is based on the combination of semi-supervised k-means method and multi-type relational data clustering. In order to achieve high performance, in the algorithm, we first analyze all kinds of relationships in data, which include intra-relationship, inter-relationship, explicit and implicit relationship; and then extend k-means clustering algorithm by seeding and new similarity measures, where attributes information, labeled data and all kinds of relationships are employed. The experimental results show the effectiveness of our method.