Jump to content

Classification Model Building Template for Spotfire Statistica® 1.0.0

1 Screenshot


This template features a typical analytical workflow for building predictive classification data mining models with Spotfire Statistica® enabling the user to change the input data source and run the whole modeling process on the new data with one click.

One of the most frequent predictive modelling types is classification which means predicting one categorical response variable with continuous or/and categorical predictor variables. Historical data with a known response is used for construction of the model, which can be afterwards used for new predictor data with an unknown response.

This template is prepared to help users quickly and simply build predictive classification models in Spotfire Statistica®. Predictive modelling is universal topic across all industries. Underlying business problems frequently lead to a simple classification task.

The model building process in this template includes following tasks:

  • splitting data into Testing and Training subsets
  • feature selection of best predictors
  • model building step for various classification methods
  • comparing the results of the models to the test data set
  • choosing the best model for future usage

For this template it is expected that input data are already in a clean form (after data preparation steps).

Release P1.0

Published: June 2018

  • Create New...