Abstract | ||
---|---|---|
This paper presents a preliminary study of an evolving clustering algorithm based on the average cluster distance - eCAD. This algorithm is mainly intended for online processing of data streams. It recursively computes the average distance between two consecutive data samples, which is included as the main measure in the evolving mechanisms. Accordingly, we have included a mechanism to automatically detect new clusters and a mechanism to merge existing clusters. This makes the algorithm autonomous and insensitive to the distribution of the data, which does not need to be prescaled or normalized. In addition, the number of clusters is not required as a predefined parameter. This algorithm could be classified as a density-based clustering algorithm and we therefore provide some comparison results with established clustering techniques. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/EAIS51927.2022.9787746 | 2022 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS) |
Keywords | DocType | ISSN |
eCAD,consecutive data samples,average distance,data streams,average cluster distance,evolving clustering algorithm | Conference | 2330-4863 |
ISBN | Citations | PageRank |
978-1-6654-3707-3 | 0 | 0.34 |
References | Authors | |
17 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Goran Andonovski | 1 | 0 | 1.01 |
Igor Skrjanc | 2 | 354 | 52.47 |