Michael Zeller

Protecting consumer privacy is key to securely providing the huge data sets required for innovations in AI analytics.

Data science algorithms are never a one-size-fits-all solution. Do you know what makes sense for your business?

The R pmml package is now able to export PMML for ensemble models via the ada and randomForest functions. In this presentation, we describe all the steps necessary to export random forest and stochastic boosting models from R into PMML and show how the PMML standard is capable of representing not only model ensembles but also any R specified treatments for missing and invalid values as well as outliers. Additional functions available to the data scientist through the R pmml package include the… Show more

The R pmml package is now able to export PMML for ensemble models via the ada and randomForest functions. In this presentation, we describe all the steps necessary to export random forest and stochastic boosting models from R into PMML and show how the PMML standard is capable of representing not only model ensembles but also any R specified treatments for missing and invalid values as well as outliers. Additional functions available to the data scientist through the R pmml package include the ability to perform data pre- and post-processing. Show less

The R pmmlTransformations package implements many of the commonly used transformation operators used by data scientists, among them the Z-transform, linear transformation, data discretization, data normalization and value mapping. The result is not only the transformed data itself but also information to represent the transformation operators in PMML format.

The Predictive Model Markup Language (PMML) is the de facto standard to represent data mining and predictive analytic models. With PMML, one can easily share a predictive solution among PMML-compliant applications and systems.

PMML as a standard has evolved significantly over the years. PMML 4.1, the language’s latest version represents a major leap forward in terms of its ability to represent data post-processing and multiple models. It also provides entirely new model elements for… Show more

The Predictive Model Markup Language (PMML) is the de facto standard to represent data mining and predictive analytic models. With PMML, one can easily share a predictive solution among PMML-compliant applications and systems.

PMML as a standard has evolved significantly over the years. PMML 4.1, the language’s latest version represents a major leap forward in terms of its ability to represent data post-processing and multiple models. It also provides entirely new model elements for supporting Scorecards and K-Nearest Neighbors. The same is no exception for PMML 4.2, currently being worked on by the Data Mining Group (DMG), the body responsible for maintaining and advancing the PMML standard. PMML 4.2 is bound to offer new elements and increased capabilities. This article describes one of such improvement. In particular, it proposes extending the existing model element for Naïve Bayes Classifiers to support continuous input fields.

The R Project is a popular choice for data miners to analyze and build predictive models. Naïve Bayes is just one of a myriad of model types supported by R. The R e1071 package provides a
naiveBayes function to build Naïve Bayes Models using categorical as well as continuous fields. The R pmml package has been recently extended to allow for the export of PMML code for objects built with the naiveBayes function. For now, it includes a PMML Extension element for continuous fields, but with the release of PMML 4.2, the support will be standardized. This article describes this process in view of our proposal to extend the current model element for Naïve Bayes Models. Show less

As the de facto standard for data mining models, the Predictive Model Markup Language (PMML) provides tremendous benefits for business, IT, and the data mining industry in general. Due to the cross-platform and vendor-independent nature of such an open-standard, it allows for predictive models to be easily moved between applications.

Although PMML has offered support for common data transformations for quite some time now, the release of PMML 4.0 in 2009 brought the support for data… Show more

As the de facto standard for data mining models, the Predictive Model Markup Language (PMML) provides tremendous benefits for business, IT, and the data mining industry in general. Due to the cross-platform and vendor-independent nature of such an open-standard, it allows for predictive models to be easily moved between applications.

Although PMML has offered support for common data transformations for quite some time now, the release of PMML 4.0 in 2009 brought the support for data pre-processing steps to a new level. As a consequence, several data mining tools and model building platforms have been adding more and more support for data pre-processing into the PMML code they export. It is no
surprise then that the same is true for R.

R has become a popular statistical platform for all things analytics. The R Project allows for a myriad of specialized packages to be installed and utilized by its users as needed. These include packages and functions for predictive analytics and model building. A package for exporting PMML out of several model types is also available. Called the pmml package, it allows for a few data pre-processing steps to be exported together with the modeling technique itself. However, a package to enable data transformations in a generic way was still missing.

This paper describes a package which intends to close this gap. The pmmlTransformations package provides R users with functions that greatly enhance the available data mining capabilities and PMML support by allowing transformations to be performed on the data before it is used for modeling. The pmmlTransformations package works in tandem with the pmml package so that data pre-processing can be represented together with the model in the resulting PMML code. Show less

Like all open standards, the Predictive Model Markup Language (PMML) enables interoperability and portability in the world of data mining and predictive analytics. This means that models developed in any environment and tool set can be deployed and used in a completely different system. Such a level of flexibility creates new opportunities for addressing exceedingly demanding business agility and performance requirements.

One of these requirements is the urgent need to apply the power of… Show more

Like all open standards, the Predictive Model Markup Language (PMML) enables interoperability and portability in the world of data mining and predictive analytics. This means that models developed in any environment and tool set can be deployed and used in a completely different system. Such a level of flexibility creates new opportunities for addressing exceedingly demanding business agility and performance requirements.

One of these requirements is the urgent need to apply the power of predictive analytics to derive reliable predictions and, hence, business decisions from vast amounts of data collected by many organizations. In this paper, we discuss how PMML enables embedding advanced predictive models directly into the database or the data warehouse, along side the actual data to be scored. More importantly, we show how we can easily take advantage of highly parallel database architectures to efficiently derive predictions from very large volumes of data. Show less

As the de facto standard for data mining models, the Predictive Model Markup Language (PMML) provides tremendous benefits for business, IT, and the data mining industry in general, since it allows for predictive models to be easily moved between applications. Due to the cross-platform and vendor-independent nature of such an open-standard, auto-generated PMML code is often represented in different versions of PMML. A tool may export PMML 2.1 and another import PMML 4.0. This problem raises the… Show more

As the de facto standard for data mining models, the Predictive Model Markup Language (PMML) provides tremendous benefits for business, IT, and the data mining industry in general, since it allows for predictive models to be easily moved between applications. Due to the cross-platform and vendor-independent nature of such an open-standard, auto-generated PMML code is often represented in different versions of PMML. A tool may export PMML 2.1 and another import PMML 4.0. This problem raises the issue of conversion. For true interoperability, PMML needs to be easily converted from one version to another.
In this paper, we describe the capabilities associated with the “PMML Converter”. This application represents a great step in the PMML path towards true interoperability in data mining. Besides converting older versions of PMML to its latest, the PMML converter checks PMML files for syntax issues and, if issues are encountered, automatically corrects them.
This paper also describes the capabilities associated with an interactive PMML-based application, the “Transformations Generator.” Auto-generated PMML code can omit important data pre-processing steps which are an integral part of a predictive solution. The Transformations Generator aims to bridge this gap by providing a graphical interface for the development and expression of data pre-processing steps in PMML. Show less

Over the past decade, we have seen tremendous interest in the application of data mining and statistical algorithms, first in research and science and, more recently, across various industries. This has translated into the development of a myriad of solutions by the data mining community that today impact scientific and business applications alike. However, even in this scenario, interoperability and open standards still lack broader adoption among data miners and modelers.

In this… Show more

Over the past decade, we have seen tremendous interest in the application of data mining and statistical algorithms, first in research and science and, more recently, across various industries. This has translated into the development of a myriad of solutions by the data mining community that today impact scientific and business applications alike. However, even in this scenario, interoperability and open standards still lack broader adoption among data miners and modelers.

In this article we highlight the use of the Predictive Model Markup Language (PMML) standard, which allows for models to be easily exchanged between analytic applications. With a focus on interoperability and PMML, we also discuss here emerging trends in cloud computing and Software as a Service, which have already started to play a critical role in promoting a more effective implementation and widespread application of predictive models.

As an illustration of how the benefits of open standards and cloud computing can be combined, we describe a predictive analytics scoring engine platform that leverages these elements to deliver an efficient deployment process for statistical models. Show less

At KDD-2009 in Paris, a panel on open standards and cloud computing addressed emerging trends for data mining applications in science and industry. This report summarizes the answers from a distinguished group of thought leaders representing key software vendors in the data mining industry.

Supporting open standards and the Predictive Model Markup Language (PMML) in particular, the panel members discuss topics regarding the adoption of prevailing standards, benefits of interoperability… Show more

At KDD-2009 in Paris, a panel on open standards and cloud computing addressed emerging trends for data mining applications in science and industry. This report summarizes the answers from a distinguished group of thought leaders representing key software vendors in the data mining industry.

Supporting open standards and the Predictive Model Markup Language (PMML) in particular, the panel members discuss topics regarding the adoption of prevailing standards, benefits of interoperability for business users, and the practical application of predictive models. We conclude with an assessment of emerging technology trends and the impact that cloud computing will have on applications as well as licensing models for the predictive analytics industry. Show less