In order to make Affinity Propagation Clustering introduced by Frey and Dueck (2007; DOI:10.1126/science.1136800) accessible to a wider audience, we ported the Matlab code published by the authors to R. The algorithms are largely analogous to the Matlab code published by Frey and Dueck. The package further provides leveraged affinity propagation and an algorithm for exemplar-based agglomerative clustering that can also be used to join clusters obtained from affinity propagation. Various plotting functions are available for analyzing clustering results.
The package is maintained by Ulrich Bodenhofer. The package itself has grown over the years in which multiple students have contributed significant parts: Johannes Palme, Chrats Melkonian, Andreas Kothmeier, and Nikola Kostic
The package can be installed from CRAN. Therefore, the the simplest way to install the package is to enter
install.packages("apcluster")into your R session. If, for what reason ever, you prefer to install the package manually, follow the instructions in the user manual.
On June 13, 2013, the maintainer of the package, Ulrich Bodenhofer, gave a webinar on the apcluster package. The webinar was hosted by the Orange County R User Group and moderated by its president, Ray DiGiacomo, Jr. The demo uses Version 1.3.2 of the package (released June 11, 2013). Link to recorded video of webinar: YouTube Video(length: 59:17)
If you encounter any issues or if you have any question that might be of interest also for other users, before writing a private message to the package developers/maintainers, please create an issue in this repository and also consider posting to the R-help Mailing List or on StackOverflow. For other matters regarding the package, please contact the package author.
If you use this package for research that is published later, you are kindly asked to cite it as follows:
Moreover, we insist that, any time you use/cite the package, you also cite the original paper in which affinity propagation has been introduced: