Prosiding
Analysis of The Accuracy of The Level Of Rainfall For Time Series Data Using Neural Network Method Based On Particle Awarm Optimized
ABSTRACT
Rainfall is an element in our daily life. Rainfall is affected the world of shipping, business, agriculture. The intensity of rainfall is different from one region to another. Data Mining and economic studies show that the method is suitable for use given by the Neural Network (NN ) with a fairly low error rate, the ability of Neural Networks in Universal Approximation has been studied by various researchers for forecasting time series data on various types of data, so that performance from Neural Network which is satisfactory in time series data forecasting. To determine the best model from each previous period, we need to optimize the weights of each relevant training data variable, so that the Neural Network method based on Particle Swarm Optimization (PSO) can be applied even though the maximum feasibility level is not certain. The biggest cost in the response to natural disasters such as floods, droughts, late planting patterns, help farmers avoid crop failures that can be caused to one's determining the types of crops grown in those days, in need right way to predict the amount of rainfall in the coming months, in order to anticipate unwanted events. Cost suppression can also be done by analyzing the changing time series patterns of rainfall data sets taken from the Romang Polong Station in Makassar from 2005-2015, which facilitates the process of analyzing the prediction of accuracy generated from the time series data sets, if implemented by combining the Particle Swarm Optimization (PSO) based on Neural Network (NN) method to find out the patterns generated from the data set. Thus saving time and operating power.
Keywords: Analysis, rainfall, time series, Neural Network, Particle Swarm Optimization.
Other Source:
https://ieeexplore.ieee.org/document/9320821/
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9320821
https://doi.org/10.1109/ICORIS50180.2020.9320821
Tidak ada salinan data
Universitas DIPA Makassar
NPP 7371142D1000002
Jln. Perintis Kemerdekaan KM.9
Telp. (0411)587194
Hotline: +6281228221994
WhatsApp Admin: +6281342092072
e-Mail: [email protected]
© 2024 — Perpustakaan UNDIPA Makassar - SLiMS