PMML: An Open Standard for Sharing Models

TLDR

PMML is an XML‑based language that has become the de‑facto standard for representing predictive and descriptive models, as well as data pre‑ and post‑processing, enabling model interchange across tools and avoiding proprietary incompatibilities. The paper aims to describe the PMML package, its overall structure and functionality, and to discuss how it can be used in R, the importance of valid PMML files, and the debate over its adoption. The PMML package, originally part of Rattle, can be accessed via Rattle’s GUI or directly in R, providing functions to export models from R to PMML and detailing the package’s structure and supported features. The authors emphasize that working with a valid PMML file is essential and highlight the ongoing debate within the data mining community regarding PMML’s widespread adoption.

Abstract

The PMML package exports a variety of predictive and descriptive models from R to the Predictive Model Markup Language (Data Mining Group, 2008). PMML is an XML-based language and has become the de-facto standard to represent not only predictive and descriptive models, but also data preand post-processing. In so doing, it allows for the interchange of models among different tools and environments, mostly avoiding proprietary issues and incompatibilities. The PMML package itself (Williams et al., 2009) was conceived at first as part of Togaware’s data mining toolkit Rattle, the R Analytical Tool To Learn Easily (Williams, 2009). Although it can easily be accessed through Rattle’s GUI, it has been separated from Rattle so that it can also be accessed directly in R. In the next section, we describe PMML and its overall structure. This is followed by a description of the functionality supported by the PMML package and how this can be used in R. We then discuss the importance of working with a valid PMML file and finish by highlighting some of the debate surrounding the adoption of PMML by the data mining community at large.