Version 0.8.15

• Fixes a regression from version 0.8.13 / 0.8.14 in categories_to_binary(), affecting the way one-hot encoding was conducted internally.

Version 0.8.13

anticlust 0.8.13

User visible changes

• anticlustering() has a new argument blocks used to sequentially apply anticlustering to different levels of a categorical variable
• categories_to_binary() no longer throws an error when a categorical variable only has 1 level

Documentation

• Extended documentation on the data input that is accepted by anticlustering()
• Extended documentation in categories_ton_binary()

Internal

• Some changes in input validation for distance matrices (which is now also explicated in the documentation to anticlustering())
• anticlust again suggests palmerpenguins (because the base R data set is not yet available on all CRAN testing machines)

Version 0.8.12

anticlust 0.8.12

• anticlustering() now has support for handling NAs
• anticlustering() now accepts factor variables as columns of the data argument x to handle categorical variables
• anticlustering() with must_link constraints now supports objective = "kplus" and objective = "variance" (for equal-sized groups)
• anticlustering() now supports method = "3phase", calling the three phase search algorithm by Yang et al., contributed by Hannah Hengelbrock (@HanneyAI)
• categories_to_binary() now treats NAs as a separate category
• anticlust now suggests packages MASS and tableone (for vignettes) and no longer suggests package palmerpenguins
• A new data set of rating norms is available, published by Brunel et al. (2025)

Version 0.8.11-1

anticlust 0.8.11-1

• Fixes some edges cases of functionality introduced in last update

Version 0.8.11

anticlust 0.8.11

• In anticlustering(), argument cannot_link can now be a vector, via 93158fb98a375, inspiration thanks to @einGlasRotwein
  • Makes the computation potentially much faster (and does not rely on a solver) if cannnot-link constraints are induced via a grouping variable.

Version 0.8.10-1

anticlust 0.8.10-1

• Minor: Fixed uncatched potential error in tests for CRAN "additional issues" via 75cb7d5760be08b0bb

Version 0.8.10

anticlust 0.8.10

• The commercial Gurobi solver is now available as backened for the exact anticlustering methods
• The documentation refers to a new anticlustering paper, available from bioRxiv

Version 0.8.9-1

anticlust 0.8.9-1

• Minor fix in three_phase_search_anticlustering() for objective = "dispersion" to address CRAN issue

Version 0.8.9

anticlust 0.8.9

User visible changes

• categories_to_binary() no longer uses dummy coding with a reference category, but instead codes each levels of a categorical variable with a separate variable (thanks to Gunnar Klau for spotting potential problems with dummy coding).
• anticlustering() has new method = "2PML", which is an improved heuristic when using must-link constraints

Version 0.8.8

anticlust 0.8.8

• three_phase_search_anticlustering() implements the three phase search algorithm by Yang et al., contributed by Hannah Hengelbrock (@HanneyAI)