Class BasicModelD


  • public class BasicModelD
    extends BasicModel
    Implements the approximation of the binomial model with the divergence for DFR. The formula used in Lucene differs slightly from the one in the original paper: to avoid underflow for small values of N and F, N is increased by 1 and F is always increased by tfn+1.

    WARNING: for terms that do not meet the expected random distribution (e.g. stopwords), this model may give poor performance, such as abnormally high or NaN scores for low tf values.

    WARNING: This API is experimental and might change in incompatible ways in the next release.
    • Constructor Detail

      • BasicModelD

        public BasicModelD()
        Sole constructor: parameter-free
    • Method Detail

      • score

        public final float score​(BasicStats stats,
                                 float tfn)
        Description copied from class: BasicModel
        Returns the informative content score.
        Specified by:
        score in class BasicModel
      • toString

        public String toString()
        Description copied from class: BasicModel
        Subclasses must override this method to return the code of the basic model formula. Refer to the original paper for the list.
        Specified by:
        toString in class BasicModel