Comparison of Model Prediction for Tile Production in Tabanan Regency with Orange Data Mining Tool

Abstract

Data can not only be saved, but the data can be processed to be a representation of new knowledge. Tile production data in Tabanan Regency is one of them, this data consists of variables that affect the results of tile production in Tabanan Regency. Tile production in Tabanan Regency is relatively declining, therefore a predictive modeling is needed to predict the results of tile production and see which variables most influence tile production. Some data mining techniques such as classification tree, AdaBost, kNN can be used to perform the prediction modeler. In this study predictions using several methods then compare the effectiveness of these methods and look for variables that are more influential on tile production using Linear Regression. Orange data mining tool is used in this study, the use of Orange Data Mining is because this tool is based on GUI so that it can be used easily and can be utilized even by ordinary people. This study produces the best value on the comparison of the methods used and the most influential variables.