Perbandingan Fungsi Optimasi Neural Network Dalam Klasifikasi Kelayakan Calon Suami

Abstract

adalah dengan menggunakan machine learning, khususnya menggunakan metode neural network . Penelitian ini bertujuan untuk menganalisa hasil klasifikasi kelayakan calon suami menggunakan metode neural network , dengan variasi kombinasi fungsi Rectified Linear Unit dan fungsi optimasi seperti Limited Memory Broyden–Fletcher–Goldfarb–Shanno Bound Constraint, Stochastic Gradient Descent dan Adative Moment Estimation . Hasil dari penelitian ini adalah bahwa kombinasi fungsi aktivasi Rectified Linear Unit dan fungsi optimasi Adaptive Moment Estimation merupakan yang Abstract The selection of a prospective husband is an important thing to consider to form a happy family. One way to help assess the suitability of a prospective husband is to use machine learning, especially using the neural network method. This study aims to analyze the results of the feasibility classification of prospective husbands using the neural network method, with variations in the combination of the Rectified Linear Unit function and optimization functions such as Limited Memory Broyden–Fletcher– Goldfarb–Shanno Bound Constraint, Stochastic Gradient Descent, and Adaptive Moment Estimation. The results of this study are that the combination of the Rectified Linear Unit activation function and the Adaptive Moment Estimation optimization function is the best, while the combination of the Rectified Linear Unit activation function and the Stochastic Gradient Descent optimization function is the worst, in terms of accuracy, precision and recall values respectively..