Prediction Operators

Prediction (Predict) operators are used to apply a particular modeling algorithm operator to a new dataset for prediction purposes.

  • Chi Square, Goodness of Fit
    Computes a Pearson's Chi Square test for goodness of fit of a distribution.
  • Chi Square, Independence Test
    Determines whether categorical columns are statistically independent of a categorical dependent variable column.
  • Classifier (DB)
    Uses any input classification model to apply a classification prediction to the input data set.
  • Classifier (HD)
    Uses any input classification model to apply a classification prediction to the input data set.
  • Collaborative Filter Predictor
    Outputs predicted ratings for products using the Collaborative Filtering model created by the trainer. Uses both the model trained by the Collaborative Filter Trainer and a data set.
  • Collaborative Filter Recommender
    Using the model trained by the Collaborative Filter Trainer, outputs recommendations for those users or products.
  • K-Means Predictor - MADlib
    The k-means predictor (MADlib) operator output is simply the assignment of the input data members to the k number of clusters, the centroids already predetermined by the K-means (MADlib) operator.
  • PCA Apply
    Uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of uncorrelated variables (principal components).
  • Predictor (DB)
    Applies an input regression, classification, or clustering model to an input data set in order to predict a value (or the highest probability value)
  • Predictor (HD)
    Applies an input regression, classification, or clustering model to an input dataset in order to predict a value (or the highest probability value).
Related concepts