Combining attribute content and label information for categorical data ensemble clustering
作者：
Highlights：
• We propose a new ensemble clustering framework that combines original data information and label information to form the information matrix.
• The new framework can be instantiated into many algorithms because of different ways of generating the set of clusterings and the final partition.
• The information matrix can be instantiated into many individuals because of different ways of combining information from original data and label.
• We form an instantiated matrix ALM considering distributions of attribute content in ensemble members and relationships among ensemble members.
摘要
•We propose a new ensemble clustering framework that combines original data information and label information to form the information matrix.•The new framework can be instantiated into many algorithms because of different ways of generating the set of clusterings and the final partition.•The information matrix can be instantiated into many individuals because of different ways of combining information from original data and label.•We form an instantiated matrix ALM considering distributions of attribute content in ensemble members and relationships among ensemble members.
论文关键词：Ensemble clustering,Information matrix,Original data information,Label information
论文评审过程：Received 3 June 2019, Revised 3 April 2020, Accepted 5 April 2020, Available online 22 April 2020, Version of Record 22 April 2020.
论文官网地址：https://doi.org/10.1016/j.amc.2020.125280
