Text
IMPLEMENTASI BOOTSTRAP SAMPLING PADA METODE ENSEMBLE BERDASARKAN METODE NAÏVE BAYES, RANDOM FOREST DAN REGRESI LOGISTIK BINER UNTUK PREDIKSI KEJADIAN HUJAN
Rain is an event of precipitation or water vapor that condenses in the atmosphere and falls to the earth's surface in the form of a liquid. Rain is highly dependent on topography and weather factors. In this case, the forecast for rain can be seen based on weather elements such as air temperature, humidity, air pressure, wind speed and so on. The purpose of this study was to predict rainfall events using the Ensemble Majority Voting method based on the Naïve Bayes method, Random Forest and Binary Logistics Regression with the implementation of Bootstrap Sampling. This study uses weather data in Australia for 2008-2017 sourced from the Kaggle website, which consists of 18 variables with a total of 145399 data entries. The level of accuracy obtained in predicting rain events using the Naïve Bayes method with the implementation of Bootstrap Sampling produces an average value of accuracy, precision, recall, and fscore of 77.94%, 50.66%, 56.37%, and 53.37%, respectively. The Random Forest method with the implementation of Bootstrap Sampling produces an average value of accuracy, precision, recall, and fscore of 82.58%, 67.04%, 43.71%, and 52.92%, respectively. The Binary Logistics Regression Method with the implementation of Bootstrap Sampling produces an average value of accuracy, precision, recall, and fscore of 82.71%, 69.93%, 39.98%, and 50.87%, respectively. The Ensemble Majority Voting method with the implementation of Bootstrap Sampling produces an average value of accuracy, precision, recall, and fscore of 82.83%, 67.51%, 44.98%, and 53.98%, respectively. The results of this study indicate that the prediction of rain events using the Ensemble Method has better accuracy than the other three methods because it has a higher accuracy value and fscore. However, the Naïve Bayes method has a better recall value and Binary Logistics Regression has a better precision value than other methods.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2207002937 | T76703 | T767032022 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available