Handwriting prediction using the support vector machine method in web-based applications

Abstract

Current technological advances have progressively encouraged individuals to use web-based applications. To determine the owner of a handwriting, one of the manual methods is generally done by people. And it can be done by a web application to find out the person’s handwriting. The Support Vector Machine (SVM) Method is a strategy of artificial intelligence used to recognize handwriting. This study intends to create a handwriting recognition framework using the Support Vector Machine (SVM) Method on a web-based application, with extraction features that are important as the basis for describing and characterizing handwritten images. The results obtained from predictions using the Support Vector Machine are by using 16 data and there are 80 tests carried out in total. The results obtained indicate that there are 25 incorrect prediction tests and 55 correct prediction tests so that the level of accuracy obtained is 68.75%.