Lodestar.Text 0.6.0. This page is frozen at that release. Read the current documentation for what main says now. A link to a decision or a migration page follows main, and leaves the archive.

RakeMetric

Which per-word score Rake sums into a phrase score.

public enum RakeMetric { DegreeToFrequencyRatio, WordDegree, WordFrequency }

Members — DegreeToFrequencyRatio sums deg(w) / freq(w): the paper's own metric, and rake-nltk's default. WordDegree sums deg(w) alone — how many words w shares a candidate with, itself included, counted once per occurrence. WordFrequency sums freq(w) alone — how often w occurs at all, ignoring how long the candidates it appears in are.

Example — the same document, three metrics, three different top scores.

using Lodestar.Text.Keywords;

string doc = "linear constraints and linear constraints";
double byDegree = new Rake(new RakeOptions { Metric = RakeMetric.WordDegree }).Extract(doc)[0].Score;
double byFrequency = new Rake(new RakeOptions { Metric = RakeMetric.WordFrequency }).Extract(doc)[0].Score;

double degreeScore = byDegree;      // => 8
double frequencyScore = byFrequency;  // => 4

Remarks — WordDegree rewards a word for standing in a long candidate — its degree grows with the run's length every time the run appears — where WordFrequency counts only how often the word itself shows up, regardless of what it stood next to. DegreeToFrequencyRatio, the default, is a compromise between the two: a word that is rare but always found in long company scores high, and one that is common but always alone does not.

Applies to — net10.0, netstandard2.0.

See also — Rake, RakeOptions, KeywordMatch.