Optimization Models

About Optimization Models

An optimization model is a supervised learning algorithm that uses input variables to select the best outcome from a set of available alternatives.

In the simplest case, an optimization problem consists of maximizing or minimizing a real function by systematically choosing input values from an allowed set and computing the value of the function.

PRIM Analysis

For more information, see the online learning platform

The objective of PRIM is to find sub-regions in the input data that have values that reach the target variable.  

To launch this model tool, select Models > PRIM analysis from the menu. .

Target variable

The target variable used in PRIM analysis can be symbolic or numerical.

Create a PRIM Analysis

The parameters for this method are defined on two tabs at the top of the page: Properties and Advanced.

On the Properties tab: 

  1. Select a Datasource from the list (if applicable).
  2. Enter a Model Name.
  3. Select a Learning set from the list.
  4. Select an variable for the Goal.
  5. Select variable(s) from the list for the Inputs.
  6. Select Goal variable.
  7. Click Save.

On the Advanced tab: 

  1. Enter an Alpha value, it is the percentage of data to be eliminated at each step. default 0.05. For more information, see Alpha
  2. Enter a Beta value,it settles the stopping criteria of the algorithm, it means the percentage of the minimum amount of data which has to be at the box, default 0.2. For more information, see Beta
  3. Enter a Target Symbol, this field is case sensitive., e.g. stable or STABLE are considered different values. For more information, see Target Symbol
  4. Choose Restrict symbol peeling, Yes or No. If selected the algorithm does not remove more records than the alpha value at each step. For more information, see Restrict symbol peeling
  5. Enter a Box Number of 1 or more. Note that once there is more than one box there is a possibility of creating an overlap. For more information, see Box Number
  6. Choose Method. “Max” stands for maximizes the target variable (average value or number of occurrences) and “min” stands for minimizes the target variable (average value or number of occurrences). For more information, see Method
  7. Choose Keep all found variables, Yes or No.  

Alpha and Beta value

It is not advisable to change the default values of Alpha and Beta.


The objective of Optimizer is to optimize a target function or model. Optimizer uses a predictive model or function variable to search an optimization based on manipulable variables and constraints. There are two optimization methods defined in DATAmaestro: Swarm & Nearest Neighbors. 

To launch this model tool, select Models > Optimizer from the menu. .

  1. Enter a Model name.
  2. Enter a Variable Prefix. Prefix to add before all variables generated by the optimization model. Default: OPT-1. 
  3. Choose the tag Target Function / Model from the list. The target is a model or function variable to optimize (required). 
  4. Select a Target options: Minimization or Maximization. For more information, see Target
  5. Enter a Target Min value. For more information, see Target Min
  6. Enter a Target Max value. For more information, see Target Max
  7. Select an optimization Method, options: Swarn or Nearest Neighbor. For more information, see Swarm
  8. Enter a Maximum iteration count, default value = 1000. For more information, Max iteration count
    1. Swarm: For Maximum number of optimization cycles before stopping (if no other stopping criteria is reached). It is a required parameter. Default: 1000.
    2. Nearest Neighbors: Number of Nearest Neighbors evaluated by the algorithm.
  9. Select a Record set from the list.
  10. Enter a Maximum record count, default value = 100.  For more information, Max record count
  11. Select manipulable variable(s) in + Add manipulable
    1. Choose a Tag from the variable list.
    2. Enter Minimum and Maximum values. Enter acceptable operating range. 
    3. Enter Step. It is the resolution of manipulable variable changes. 
    4. To delete the tag, click on the trash icon
  12. Select constraint variable(s) in + Add constraint.


    Select models or function variable that are based on Manipulable variables to check that changes to manipulable variables respect process constraints.

    1. Choose a Tag from variable list.
    2. Enter Minimum and Maximum values. Enter accepting operating range. 
    3. To delete the tag, click on the trash icon
  13. Click Save.