Abstract
Data has become an integral part of our society in the past years, arriving faster and in larger quantities than before. Traditional clustering algorithms rely on the availability of entire datasets to model them correctly and efficiently. Such requirements are not possible in the data stream clustering scenario, where data arrives and needs to be analyzed continuously. This paper proposes a novel evolutionary clustering algorithm, entitled EvolveCluster, capable of modeling evolving data streams. We compare EvolveCluster against two other evolutionary clustering algorithms, PivotBiCluster and Split-Merge Evolutionary Clustering, by conducting experiments on three different datasets. Furthermore, we perform additional experiments on EvolveCluster to further evaluate its capabilities on clustering evolving data streams. Our results show that EvolveCluster manages to capture evolving data stream behaviors and adapts accordingly.
Highlights
In recent years, data has become an integral part of our daily lives
1 https://github.com/christiannordahl/EvolveCluster the EvolveCluster and Split-Merge Clustering algorithms are more proficient than PivotBiCluster in identifying new clusters when they arrive in the data stream
It is interesting to notice that Split-Merge Clustering fully follows the true clustering of the data points, up to the point that even data points that are overlapping into another cluster is correctly classified
Summary
Data has become an integral part of our daily lives. Due to advances in hardware infrastructures, there are endless possibilities available to collect any type of data at a rapid pace. (Bifet et al 2010b) These data streams are endless information sources that arrive in a timely fashion. The results from the unsupervised learning algorithms can be used directly for analysis or as an intermediary step to gain an understanding of the data. One of the branches of unsupervised learning is the task of clustering analysis. Clustering algorithms are designed to identify an underlying structure of data and use the detected relationships within the structure to group the data points into distinct groups. These algorithms usually decide upon themselves how to divide the data into subgroups, an unsupervised approach to increase knowledge about the data. This study focuses on partitioning algorithms due to the proposed evolutionary clustering algorithms characteristics (see Sect. 4)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
