This package contains nonparametric kernel methods for calculating pairwise distances between mixed-type observations. These methods can be used in any distance based algorithm, with emphasis placed on usage in clustering or classification applications. Descriptions of the implementation of these methods can be found in Ghashti (2024) and Ghashti and Thompson (2024).
John R.J. Thompson <john.thompson@ubc.ca>, Jesse S. Ghashti <jesse.ghashti@ubc.ca>
Maintainer: John R.J. Thompson <john.thompson@ubc.ca>
We would like to acknowledge funding support from the University of British Columbia Aspire Fund (UBC:www.ok.ubc.ca/). We also acknowledge support from the Natural Sciences and Engineering Research Council of Canada (NSERC).
This package contains two functions for pairwise distance calculations of mixed-type data based on two different methods. Kernel methods also require variable-specific bandwidths, with two additional functions for the bandwidth specification methods. Additionally, this package contains a function methods for mixed-type data generation.
Ghashti, J.S. (2024), “Similarity Maximization and Shrinkage Approach in Kernel Metric Learning for Clustering Mixed-type Data (T)”, University of British Columbia. <https://dx.doi.org/10.14288/1.044397>
Ghashti, J.S. and J.R.J Thompson (2024), “Mixed-type Distance Shrinkage and Selection for Clustering via Kernel Metric Learning”. Journal of Classification, Accepted.