_textSimilarity()

in Readability.js [981:996]


  _textSimilarity(textA, textB) {
    var tokensA = textA
      .toLowerCase()
      .split(this.REGEXPS.tokenize)
      .filter(Boolean);
    var tokensB = textB
      .toLowerCase()
      .split(this.REGEXPS.tokenize)
      .filter(Boolean);
    if (!tokensA.length || !tokensB.length) {
      return 0;
    }
    var uniqTokensB = tokensB.filter(token => !tokensA.includes(token));
    var distanceB = uniqTokensB.join(" ").length / tokensB.join(" ").length;
    return 1 - distanceB;
  },