Given a set of multivariate time series, the problem of clustering such data is concerned with the discovering of inherent groupings of the data according to how similar or dissimilar the time series are to each other...
详细信息
ISBN:
(纸本)9783319131863;9783319131856
Given a set of multivariate time series, the problem of clustering such data is concerned with the discovering of inherent groupings of the data according to how similar or dissimilar the time series are to each other. Existing time series clusteringalgorithms can divide into three types, raw-based, feature-based and model-based. In this paper, a model-based multivariate time series clusteringalgorithm is proposed and its tasks in several steps: (i) data transformation, (ii) discovering time series temporal patterns using confidence value to represent the relationship between different variables, (iii) clustering of multivariate time series based on the degree of patterns discovering in (ii). For evaluate performance of proposed algorithm, the proposed algorithm is tested with both synthetic data and real data. The result shows that it can be promising algorithm for multivariate time series clustering.
In observational studies, unbalanced observed covariates between treatment groups often cause biased inferences on the estimation of treatment effects. Recently, generalized propensity score (GPS) has been proposed to...
详细信息
In observational studies, unbalanced observed covariates between treatment groups often cause biased inferences on the estimation of treatment effects. Recently, generalized propensity score (GPS) has been proposed to overcome this problem;however, a practical technique to apply the GPS is lacking. This study demonstrates how clusteringalgorithms can be used to group similar subjects based on transformed GPS. We compare four popular clusteringalgorithms: k-means clustering (KMC), model-basedclustering, fuzzy c-means clustering and partitioning around medoids based on the following three criteria: average dissimilarity between subjects within clusters, average Dunn index and average silhouette width under four various covariate scenarios. Simulation studies show that the KMC algorithm has overall better performance compared with the other three clusteringalgorithms. Therefore, we recommend using the KMC algorithm to group similar subjects based on the transformed GPS.
暂无评论