Comparative Analysis of BPNN and LVQ for Sundanese Character Recognition

The Sundanese script (Aksara Sunda), an essential part of Sundanese cultural heritage, has been used since the 14th century AD. However, recognizing handwritten Sundanese characters remains challenging due to variations in individual writing styles. This study compares the performance of Backpropagation Neural Network (BPNN) and Learning Vector Quantization (LVQ) for recognizing handwritten Sundanese vowel (Swara) characters. A dataset was collected from 15 individuals, each writing seven Sundanese vowel characters, which were then used for training and testing the recognition models. Experimental results show that BPNN outperforms LVQ, achieving a higher classification accuracy (95.23%), lower Mean Squared Error (MSE), and faster convergence compared to LVQ, which reached a maximum accuracy of 66.66%. Additionally, BPNN demonstrated better generalization and robustness. At the same time, LVQ was highly sensitive to learning rate variations, leading to unstable accuracy and slower training times. The findings highlight that BPNN is a more effective model for Sundanese script recognition, providing a reliable approach for preserving and digitizing traditional scripts. Future research should explore hybrid models, deep learning approaches, and larger datasets to enhance recognition accuracy and system robustness.