Conversion of adverse data corpus to shrewd output using sampling metrics

Visual Computing for Industry, Biomedicine, and Art

Table 3 Results of classification with the imbalanced dataset

Classifier	Accuracy	Classes	Precision	Recall	F-Measure	Confusion matrix
Naïve bayes	84.77%	Low	0.659	0.750	0.701	a b < −- classified as 27 9 \| a = Low 14,101 \| b = High
		High	0.918	0.878	0.898
		Average	0.856	0.848	0.851
Multilayer perceptron	80.79%	Low	0.600	0.583	0.592	a b < −- classified as 21 15 \| a = Low 14,101 \| b = High
		High	0.871	0.878	0.874
		Average	0.806	0.808	0.807
SVM	89.40%	Low	0.833	0.694	0.758	a b < −- classified as 25 11 \| a = Low 5110 \| b = High
		High	0.909	0.957	0.932
		Average	0.891	0.894	0.891
IBk	78.81%	Low	0.559	0.528	0.543	a b < −- classified as 19 17 \| a = Low 15,100 \| b = High
		High	0.855	0.870	0.862
		Average	0.784	0.788	0.786
Random forest	86.09%	Low	0.727	0.667	0.696	a b < −- classified as 24 12 \| a = Low 9106 \| b = High
		High	0.898	0.922	0.910
		Average	0.858	0.861	0.859