Description:
Preprocessed Dataset in Weka using Visualization, Normalization and Discretization.
Used Wrapper and Filter method for attribute selection.
Used Naive Bayes, Bayesian net and Support vector machine (SMO) to check the comparative results across the three methods for testing dataset
Used Adboost, Bagging and stacking on top of decision tree for model enhancement.
Key Results:
- Data Preprocessing Using Weka
Visualization Normalization
Discretization
- Advanced Modeling in Weka
Naive Bayes Algorithm:
a b <– classified as
1599 228 | a = 0 Correctly Classified Instances 1721 81.5253 %
162 122 | b = 1 Incorrectly Classified Instances 390 18.4747 %
BayesNet Algorithm:
a b <– classified as
1622 205 | a = 0 Correctly Classified Instances 1750 82.8991 %
156 128 | b = 1 Incorrectly Classified Instances 361 17.1009 %
SMO Algorithm
a b <– classified as
1827 0 | a = 0 Correctly Classified Instances 1827 86.5467 %
284 0 | b = 1 Incorrectly Classified Instances 284 13.4533 % Over fitting
Among three, the least cost of misclassification is of BayesNet Algorithm