<![CDATA[This study was made to classify the types of children's reading books based image on the on the cover. The types of books used in this research are fairy tales, educational books and comics. The problem that occurs is that there are many types of reading books so that there are no errors in identifying children's literature, then classification is carried out. Uses approximately 1002 image data. This study uses the method Convolutional Neural Network (CNN). The purpose of this study was to classify the types of children's reading books to suit the age of the reader. It is hoped that the research objectives can make it easier for parents to find books that are appropriate for their child's age. Convolutional Neural Network (CNN) is a method deep learning that is usually used to process data in the form of images. The research stages start from literature study, data collection, data processing, needs analysis, design, implementation and testing. The collection uses several methods, namely: literature study, documentation method, interview and questionnaire method. The design is carried out from input image data, followed by calculating the accuracy and classification process which results in image classification. Implementation using the Python programming language. Evaluation of the performance of testing the accuracy value using the confusion matrix. The result of the research is a system that can classify the types of children's reading books using 70% training data and 30% test data. With an accuracy rate above 80%.]]>
Copyrights © 2022
