00:02
A, ensemble classifiers has been successful in generating high accuracies because they combine multiple individual classifiers to make predictions.
00:22
Each individual classifier may have its own strength and weakness, but by combining them, the ensemble classifier have leveraged the strength of each individual classifier and mitigate their weakness.
00:35
This can lead to improved overall accuracy.
00:37
One specific example of this is random forest.
00:41
So random forest combines that combines multiple decision trees where each tree is trained on a random subset of a training data and features.
00:59
This prediction of the ensemble classifier is determined by aggregating the prediction of all the individual decision trees.
01:08
The random forest has shown a highly accurate in various applications such as image classification, text classification and medical diagnosis.
01:19
If the training examples are linearly separable support vector machine svm can find multiple decision boundaries that separate positive from negative data points.
01:33
The number of decision boundaries depends upon number of support vectors, which are the data points closest to the decision boundary.
01:42
So the svm algorithm calculates decision boundary that maximizes the margin between the support vectors of the negative and positive classes...