With the rapid development of data collection and storage technologies, the volume of data is getting so enormous for collection and analysis in a reasonable amount of time. Only a small fraction of the original data could be contained in the databases or data warehouses. Traditional clustering approaches are recognized as an indispensable solution to extract useful knowledge from data. However, existing conventional clustering methods all lack of robustness and computation efficiency when applied on massive data. In this work, we have made several efforts to better address the above problems with novel techniques of automatic window initialization, distribution density threshold, and window traversal based on distribution density.
CITATION STYLE
Xu, X., Zhang, G., & Wu, W. (2015). A fast distribution-based clustering algorithm for massive data. In Lecture Notes in Electrical Engineering (Vol. 355, pp. 323–330). Springer Verlag. https://doi.org/10.1007/978-3-319-11104-9_38
Mendeley helps you to discover research relevant for your work.