<![CDATA[Handwriting is a biometric characteristic because each person has a unique handwriting pattern. This uniqueness can be utilized as a biometric identity. Handwriting pattern recognition is one of the important fields in document analysis to biometric authentication. This research explores the implementation of K-Nearest Neighbors (K-NN) and Support Vector Machine (SVM) algorithms in the context of handwriting pattern recognition. In addition, this research incorporates digital image processing technology by utilizing feature extraction using Gray-Level Co-occurrence Matrix (GLCM). This process involves taking handwriting samples, digitizing them into digital images, and utilizing GLCM to extract texture features. These features play an important role in capturing the unique characteristics of each handwriting pattern. This research was conducted because handwriting has a wide implementation in various fields. In the field of data security, handwriting recognition can be used for data verification in financial transactions and official documents. A comparison of the K-NN and SVM algorithms was conducted to determine the most effective and efficient algorithm in handwriting pattern recognition. These two algorithms are very popular and often used in classification. By comparing these two algorithms, this research aims to evaluate and compare the performance of two classification algorithms in handwriting pattern recognition so as to provide recommendations for implementation in handwriting pattern recognition. The main focus of this research is to investigate the effectiveness and accuracy of the K-NN and SVM algorithms in recognizing and classifying handwriting. K-NN algorithm produces the highest accuracy value of 82.11%, while the SVM algorithm produces the highest accuracy value of 83.87%, so that the SVM algorithm becomes the best algorithm in the classification of handwriting pattern recognition.]]>
Copyrights © 2024
