Data Miner Recipes (DMR) Overview

You can build advanced analytic models to relate one or more target (dependent) quantities to a number of input (independent) predictor variables using Data Miner Recipes (DMR).

The target variables are continuous or categorical.
  • Continuous target variables are usually associated with regression tasks
  • Categorical variables are used in classification problems
DMR handles both types of variables and also helps building predictive models for tackling regression and classification problems.

DMR is a complete solution that makes the process of predictive model building a systematic and step-by-step process.

Model building in DMR includes:
  1. Data cleaning, variable transformation, dimensionality reduction, eliminating redundancy.
  2. Build various predictive models; neural networks, support vector machine, trees.
  3. Model evaluation.
  4. Model deployment to score (predict) against new data.

In addition to a recipe-like user interface for building predictive models, DMR also supports the off-loading of computationally demanding tasks. With DMR, you can save projections and reload them in the future for further deployment.