Building a Classification Model with Spark - Machine Learning with Spark - page 181

}
val metrics = new
BinaryClassificationMetrics(scoreAndLabels)
(model.getClass.getSimpleName, metrics.areaUnderPR,
metrics.areaUnderROC)
}
val allMetrics = metrics ++ nbMetrics ++ dtMetrics
allMetrics.foreach{ case (m, pr, roc) =>
println(f"$m, Area under PR: ${pr * 100.0}%2.4f%%, Area
under ROC: ${roc * 100.0}%2.4f%%")
}
Your output will look similar to the one here:
LogisticRegressionModel, Area under PR: 75.6759%, Area
under ROC: 50.1418%
SVMModel, Area under PR: 75.6759%, Area under ROC: 50.1418%
NaiveBayesModel, Area under PR: 68.0851%, Area under ROC:
58.3559%
DecisionTreeModel, Area under PR: 74.3081%, Area under ROC:
64.8837%
We can see that all models achieve broadly similar results for the average precision met-
ric.
Logistic regression and SVM achieve results of around 0.5 for AUC. This indicates that
they do no better than random chance! Our naïve Bayes and decision tree models fare a
little better, achieving an AUC of 0.58 and 0.65, respectively. Still, this is not a very good
result in terms of binary classification performance.
Note
While we don't cover multiclass classification here, MLlib provides a similar evaluation
class called MulticlassMetrics , which provides averaged versions of many com-
mon metrics.