A new Growing Neural Gas for clustering data streams

• Published in: Neural networks Vol. 78; pp. 36 - 50
• Main Authors: Ghesmoune, Mohammed, Lebbah, Mustapha, Azzag, Hanene
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01-06-2016
• Subjects: Algorithms; Cluster Analysis; Clustering; Clusters; Data stream clustering; Data transmission; Databases, Factual - trends; Datasets as Topic - trends; GNG; Humans; Memory; Neural networks; Neural Networks (Computer); Partitioning; Reservoirs; Statistics as Topic - trends; Tasks; Topological structure; Topological structure; GNG; Data stream clustering
• ISSN: 0893-6080; 1879-2782
• DOI: 10.1016/j.neunet.2016.02.003

Clustering data streams is becoming the most efficient way to cluster a massive dataset. This task requires a process capable of partitioning observations continuously with restrictions of memory and time. In this paper we present a new algorithm, called G-Stream, for clustering data streams by making one pass over the data. G-Stream is based on growing neural gas, that allows us to discover clusters of arbitrary shapes without any assumptions on the number of clusters. By using a reservoir, and applying a fading function, the quality of clustering is improved. The performance of the proposed algorithm is evaluated on public datasets.