Font Size:
A Topological Clustering on Evolutionary Data
Last modified: 2023-07-06
Abstract
This paper propose a Topological approach of Clustering on Evolutionary Data (TCED), clustering that results from exploratory methods for the joint analysis of several data tables; more specifically, methods that can be applied to temporal data.
This approach is based on the notion of neighborhood graphs in an evolutionary data context. It makes it possible to simultaneously explore several tables of data collected at different times on the same individual rows, even in cases where the variables are different in the tables considered.
TCED analyzes the structure of correlations or associations observed between the variables according to their type in each table.
This approach is illustrated on real data, the results are compared with those resulting from the clustering on the significant factors of the multiple factorial analysis (MFA).
This approach is based on the notion of neighborhood graphs in an evolutionary data context. It makes it possible to simultaneously explore several tables of data collected at different times on the same individual rows, even in cases where the variables are different in the tables considered.
TCED analyzes the structure of correlations or associations observed between the variables according to their type in each table.
This approach is illustrated on real data, the results are compared with those resulting from the clustering on the significant factors of the multiple factorial analysis (MFA).