Open Access: An efficient k-modes algorithm for clustering categorical datasets

Each week, we select a recently published Open Access article to feature. This week’s article comes from Statistical Analysis and Data Mining and proposes a new method for the implementation of k-modes. 

The article’s abstract is given below, with the full article available to read here.

Dorman, K. S.Maitra, R.An efficient k-modes algorithm for clustering categorical datasetsStat. Anal. Data Min.: ASA Data Sci. J.. (2021), 1– 15https://doi.org/10.1002/sam.11546
 
Mining clusters from data is an important endeavor in many applications. The k-means method is a popular, efficient, and distribution-free approach for clustering numerical-valued data, but does not apply for categorical-valued observations. The k-modes method addresses this lacuna by replacing the Euclidean with the Hamming distance and the means with the modes in the k-means objective function. We provide a novel, computationally efficient implementation of k-modes, called Optimal Transfer Quick Transfer (OTQT). We prove that OTQT finds updates to improve the objective function that are undetectable to existing k-modes algorithms. Although slightly slower per iteration due to algorithmic complexity, OTQT is always more accurate and almost always faster (and only barely slower on some datasets) to the final optimum. Thus, we recommend OTQT as the preferred, default algorithm for k-modes optimization.
 
More Details