Analysis of Classification Technique for Prediction of Damages Levels in Building-Structures

Abstract

This article proposes various classification reviews for predicting the damage levelsin the Building dataset. Some of theseClassification Algorithmsare Support Vector Machine (SVM), Artificial NeuralNetwork (ANN),K-nearest Neighbour (KNN), Naïve Bayes (NB), Decision Tree (DT) algorithms. The five -fold cross validation assessment of classification algorithm applied on the Building dataset to predict the damagelevels in thebuilding.From ourinvestigation, it is observed that theDecision Tree (DT) gets higheraccuracywhen compare to SVM, NBand ANN algorithms for prediction of Building damage levels. Hence DT algorithm exactly suitable for Building damage prediction based on the observation in the dataset. Finally this study helps investigators for selecting the appropriate approach for predicting the damage levels of the Building. The experiments are conducted on WEKA machine learning tool. The measures such as accuracy, precision, Recall, and F-measure are calculated for above classification algorithms.