Papers › Batch Self Organizing maps for distributional data using adaptive distances
Batch Self Organizing maps for distributional data using adaptive distances
Antonio Irpino, Francisco De Carvalho, Rosanna Verde, Antonio Balzanella
The archive published only this paper's code-link row. Authors, date and abstract are from arXiv's metadata (CC0), read from the Kaggle arXiv metadata snapshot of 2026-09-12 where its title matched the archive's; the title is the archive's.
The paper deals with a Batch Self Organizing Map algorithm (DBSOM) for data described by distributional-valued variables. This kind of variables is characterized to take as values one-dimensional probability or frequency distributions on a numeric support. The objective function optimized in the algorithm depends on the choice of the distance measure. According to the nature of the date, the L₂ Wasserstein distance is proposed as one of the most suitable metrics to compare distributions. It is widely used in several contexts of analysis of distributional data. Conventional batch SOM algorithms consider that all variables are equally important for the training of the SOM. However, it is well known that some variables are less relevant than others for this task. In order to take into account the different contribution of the variables we propose an adaptive version of the DBSOM algorithm that tackles this problem with an additional step: a relevance weight is automatically learned for each distributional-valued variable. Moreover, since the L₂ Wasserstein distance allows a decomposition into two components: one related to the means and one related to the size and shape of the distributions, also relevance weights are automatically learned for each of the measurement components to emphasize the importance of the different estimated parameters of the distributions. Examples of real and synthetic datasets of distributional data illustrate the usefulness of the proposed DBSOM algorithms.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections