<![CDATA[The use of learning media in the teaching and learning process is an effort to increase the effectiveness and quality of the learning process. However, the need for learning media is not compatible with the number of learning media made by the teacher himself. One of the factors that causes it is the teacher's workload which is quite a lot so that the teacher does not have enough time to make his own learning media. This study aims to measure the extent of the correlation between the teacher's workload and the amount of instructional media that the teacher himself made. Artificial Neural Network with Backpropagation method is a tool that can be used to solve complex problems, one of which is to measure the level of correlation. The ability of an Artificial Neural Network with the Backpropagation method to adapt to changes that occur in the input and output values makes the prediction accuracy quite high. The teacher's workload variables used are the number of face-to-face hours of even and odd semesters, additional assignments (deputy principal/head of laboratory), homeroom teacher, and extracurricular coaches. The target used is the number of learning media made by the teacher himself. The data used in this study were taken from the workload of teachers at SMAN 4 Payakumbuh in 2022. The architectural patterns used are 5-4-1, 5-5-1, 5-7-1, 5-10-1, and 5- 12-1. From the test results with the Matlab R2013a software, the best pattern was obtained, namely the 5-12-1 pattern with an MSE value of 0.1001, a MAPE of 2.11, and a data accuracy of 97.89%. From the results of the training and testing, it was concluded that the correlation between the teacher's workload and the amount of self-made learning media is very low or not closely related.]]>
Copyrights © 2023
