The rates vary by length of haul, size of shipment, and the product's classification (a number that reflects the ease with which the carrier can handle the product; for example, gravel has a low classification number, bees in hives have a high classification number).
However, the random forest yields a slightly low classification accuracy on the proposed method.
But, the random forest classifier shows a slightly low classification accuracy on NSL-KDD dataset.
However, the proposed method shows a slightly low classification accuracy with random forest and SVM classifiers.
But, we found a bit low classification accuracy with SVM classifier for our method.
But, decision trees, random forest and KNN show low classification accuracy on the proposed method.
However, KNN yields a slightly low classification accuracy for the proposed method.
The randomized rules generated also have low classification abilities on the original dataset.
The low classification error indicates the high level of information gain.
However, the original multiple-instance learning has the weaknesses of low classification effectiveness and real-time ability.
