Penerapan Pl/SQL Untuk Menentukan Pola Data Latih Pada Aplikasi Berbasis Neural Network

Abstract

Stock stability in business activities is an important component. One way to maintain the stability of the stock of durian fruit is to estimate the amount of consumer demand for durian fruit. Predicting the amount of durian fruit customers need can be done by looking at the patterns formed from sales transactions. However, if the prediction is done manually, it will take a long time and be prone to errors. The solution to overcome this problem is the need for a tool in the form of software in which there is already an accurate calculation method. One method that can be used to calculate customer durian needs is the Adaline artificial neural network. And the training data used is durian fruit sales data. Training data is divided into 2 groups, namely input data and target data. The focus of this research is the pl/sql design to determine the data input pattern automatically for the application of predicting the number of customer's durian fruit needs. The types of pl/sql used in this study are procedure and function