Performance recognition on handwritten characters using novel support vector machine and compare with logistic regression.

Autor: Kummitha, Vengala Reddy, Kumar, S. Magesh
Předmět:
Zdroj: AIP Conference Proceedings; 2024, Vol. 2853 Issue 1, p1-7, 7p
Abstrakt: The goal of this study is to use Support Vector Machines (SVMs) to recognise handwritten characters and compare their predictive accuracy to that of Logistic Regression. In this paper, we explore the application of machine learning algorithms like Support Vector Machine and Logistic Regression to the problem of character recognition. For a 0.05 level of statistical significance, we need 15 samples out of a total of 30. The SVM algorithm achieves a higher recognition rate (96.58 percent) than the Logistic Regression algorithm (87.43 percent) when both are used for character recognition. A p-value of 0.005 indicates a statistically significant difference between the study groups. Support Vector Machine was found to have better prediction accuracy than the Logistic Regression algorithm in this study of character recognition. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index