Penerapan Data Mining Dengan Algoritma Neural Network(Backpropagation) Untuk Prediksi Pemakaian Listrik
Total View This Week1
Total View This Week1
Institusion
STMIK Bumi Gora
Author
Candra, M. Ade
Subject
QA75 Electronic computers. Computer science
Datestamp
2021-03-08 06:41:36
Abstract :
Electricity has become a necessity that cannot be separated from the activities carried out by humans every day. The role of electricity is not only a secondary need, but has turned into a primary need because without electricity all activities carried out will be very obstructing. This human dependence on electricity creates bad habits. For this reason, the authors feel the need to conduct research by analyzing the backpropagation method. This research uses data mining stages, namely data collection, pre-processing, such as data transformation. The third stage is making a model using the backpropagation method and the last stage is the evaluation of the results by measuring the performance of the backpropagation algorithm based on RMSE. The data used is data on electricity consumption for the Mataram, Sumbawa, and Bima areas in the 2008-2018 timeframe, where the data processing uses the backpropagation method. The test results in this study get the best performance on the 6 hidden layer architecture and 0.4 learning rate by getting an RMSE value of 0.203424 with an accuracy rate of 84.08%. Thus, the prediction of electricity usage using the backpropagation method can be applied because it has good performance.
Electricity has become a necessity that cannot be separated from the activities carried out by humans every day. The role of electricity is not only a secondary need, but has turned into a primary need because without electricity all activities carried out will be very obstructing. This human dependence on electricity creates bad habits. For this reason, the authors feel the need to conduct research by analyzing the backpropagation method. This research uses data mining stages, namely data collection, pre-processing, such as data transformation. The third stage is making a model using the backpropagation method and the last stage is the evaluation of the results by measuring the performance of the backpropagation algorithm based on RMSE. The data used is data on electricity consumption for the Mataram, Sumbawa, and Bima areas in the 2008-2018 timeframe, where the data processing uses the backpropagation method. The test results in this study get the best performance on the 6 hidden layer architecture and 0.4 learning rate by getting an RMSE value of 0.203424 with an accuracy rate of 84.08%. Thus, the prediction of electricity usage using the backpropagation method can be applied because it has good performance.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
