CLASSIFICATION OF BABYLONIAN NUMBERS USING CONVOLUTIONAL NEURAL NETWORKS

Abstract

This study presents a novel approach for classifying Babylonian numerals through the use of Convolutional Neural Networks (CNNs). The method combines structural feature analysis focusing on vertical and horizontal angles of cuneiform symbols with the capabilities of deep learning. By applying CNN architectures, the proposed system achieves a high level of accuracy in recognizing and interpreting these ancient numerical forms. Beyond offering an efficient classification tool, our research contributes to the preservation and analysis of historical mathematical texts. Experimental results show a classification accuracy of 98.33%, demonstrating the potential of deep learning methods in the study and safeguarding of ancient cultural heritage.