New features in FormuSense allow users to handle non-linearities with data transformations.

While the multivariate models developed in FormuSense are inherently linear, the “Transformations Release” adds features that may improve model fit when non-linearities exist in the dataset.

Users may configure custom data transformations, all possible transformations, or let FormuSense intelligently suggest a controlled number of transformations. All transformations generated become new variables available for inclusion in PCA and PLS modeling.

Start Your Trial


Download a copy of FormuSense from the FormuSense homepage, and activate your 30-day trial today.
 

Attend a Demo or Workshop

Register for an upcoming demo (typically 45-min) or an interactive product development workshop (2-hours) with ProSensus by browsing the FormuSense events page.

Release Notes

Below is an expandable summary of new features added and changes implemented in FormuSense as part of the version released in July 2024. This version is referred to as the Transformations Release, but also includes other minor updates.

What’s New

Transforms
    • Configuration:
      • Generate data transformations to handle non-linearities and interactions with options for equations, property selection, and variable ranges.
      • Transformations can be manually configured. Users may select a data block of interest, then specify which specific variable(s) and transformation equation(s) to apply.
      • Users also have the option to apply all possible transformations to all variables in a data block.
      • Alternatively, FormuSense can suggest data transformations to the user, by evaluating all possible options and selecting those that meet configurable settings (such as minimum correlation to a quality variable, and maximum percent missing data).
  • Generated Transforms:
    • Once the transformations have been configured and generated, users may view line plots of all transformations and a bar plot of the percent missing.

What’s Changed

Ratios
    • Ability to copy plots to clipboard.
Mixture Properties
    • Generated Properties:
      • When the mixture property configuration would result in repeated mixture property names, the ingredient class name is appended (where needed) to generate unique variable names.
      • Ability to copy plots to clipboard.
Modeling
    • Setup:
      • Ability to include transform variables in PCA and PLS models.
      • Ability to copy plots to clipboard.
    • Analysis:
      • Ability to sort contribution plots, coefficient plots, and variable importance plots in descending order.
      • View a ta vs ua plot. This plot can be used to identify a well-fit model and the existence of non-linearities.
Simulation
    • Results:
      • View transform values resultant of simulation setup.
      • Ability to copy plots to clipboard.
Optimization
    • Setup:
      • Ability to set limits on transforms.
    • Results:
      • View transform values resultant of optimization setup.
      • Ability to copy plots to clipboard.
Export
    • Export transforms.