Score3.scala

private def countSimilarity(target: Array[String], rdd: RDD[(String, Seq[String])]): RDD[(String, Double, Seq[String])] = {
  rdd.map {
    case (author, words) =>
      val matching = (for {
        w <- words
        t <- target
        if t.length > 3 && w.length > 3 //remove short words
        score = scoreSimilarity(t, w)
        if score > 0.65d //remove accidental, short matches
      } yield (score, w)).distinct //remove duplicated results

      val score = Math.min(matching.map(_._1).sum, target.length)
      val wordsMatched = matching.map(_._2)

      (author, score, wordsMatched)
  }.sortBy(_._2, false)
}