Transfer Clustering Based on Gaussian Mixture Model
Rongrong Wang, Jin Zhou, Xiangdao Liu, Shiyuan Han, Lin Wang, Yuehui Chen · 2019
Gaussian mixture model is a helpful method for data mining. However, when the data is scarce, the traditional clustering algorithm based on Gaussian mixture model is not effective anymore. To solve this issue, this paper presents a novel transfer clustering algorithm based on Gaussian mixture model, which utilizes the information of the data in the source domain to impact to cluster the data of the target domain. In this method, traditional Gaussian mixture model is first applied in the source domain to extract the mean and the covariance of each Gaussian distribution. Then the data in the target domain is clustered under the influence of the extracted mean and covariance from the source domain. Experiments on synthetic datasets demonstrate the efficiency of the proposed algorithm compared with the traditional Gaussian mixture model.