PERBANDINGAN KLASIFIKASI PECEMARAN AIR SUNGAI DENGAN METODE BACKPROPAGATION DAN NAÏVE BAYES

Intisari / Abstract

Backpropagation and Naïve Bayes algorithms can be used to construct a classification model from a set of input data. Each algorithm has a different approach in classification. This study will try to compare two classification algorithms, namely Backpropagation and Naive Bayes. Both algorithms will be compared the value of accuracy in the classification of river water pollution. The architecture is built using a dataset of 150 data with 22 input neurons and 3 output neurons. The test is done using vallidation matrix with 4 different fold and split conditions. The results obtained Naive Bayes algorithm has a better average accuracy of 72.79% compared with Backpropagation algorithm which only has an accuracy of 64.75%.