A new sparse representation learning of complex data: Application to dynamic clustering of web navigation
作者：
Highlights：
• We propose prototype-based clustering algorithms for relational data, i.e. data described by their similarities.
• The algorithms are adapted to any kind of objects, the prototypes being computed using the Barycentric Coordinate system.
• The experimental results confirm the effectiveness of the algorithm to deal with the complexity of massive data in terms of computation time and memory.
• We present an application to analyze the structure and dynamics of user’s interest on the Internet.
摘要
•We propose prototype-based clustering algorithms for relational data, i.e. data described by their similarities.•The algorithms are adapted to any kind of objects, the prototypes being computed using the Barycentric Coordinate system.•The experimental results confirm the effectiveness of the algorithm to deal with the complexity of massive data in terms of computation time and memory.•We present an application to analyze the structure and dynamics of user’s interest on the Internet.
论文关键词：Clustering,Relational data,Barycentric coordinates,Data stream
论文评审过程：Received 28 May 2018, Revised 15 November 2018, Accepted 22 February 2019, Available online 27 February 2019, Version of Record 6 March 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.02.020
