A New Scheme to Visualize Clusters Model in Data Mining

摘要

This paper presents the design and implementation of a new scheme to visualize clusters model called IVCM, in the context of a data-mining process. The visualization of a cluster model becomes complex when the dataset is high volume, density and dimensionality. The design of IVCM scheme is based on four characteristics: interactive visualization, combination of models, ad-hoc graphics artifacts, and use metrics. The objective of this scheme is to contribute to the analysis and understanding of a clustering model. Metrics considered in this proposed scheme, allow comparison instances of different clusters, which in turn helps to understand how groups are composed. Through the implementation of a web visual environment that meets the characteristics defined in IVCM, and an online assessment of 23 users, positive results on the usefulness of this new visualization scheme are achieved.