Analisis Grafologi Berdasarkan Huruf a dan t Menggunakan Jaringan Syaraf Tiruan Propagasi Balik Standar

Abstract

Graphology or handwriting analysis is a method of identifying, analysing, evaluating, and understanding personality through the strokes and patterns revealed by handwriting. People who can analyze handwriting called graphologist. Graphologist can have a subjective assesment. Different graphologist can analyze the same handwriting but give different results. In addition, the accuracy of handwriting analysis depend on the graphologists ability. A system that can recognize handwrititng patterns is required to overcome these problems. Identification system which is implemented in this research uses Backpropagation Neural Network (BPNN). This research used 135 images of letter a and 150 images of letter t. This research is divided into three parts which are determining optimal combination of BPNN parameter, identifying personality based on letter a, and identifying personality based on letter t then classifies them into one of three character classes available. The results of this research is that the system has 98.15% accuracy for letter a and 73.33% for letter t. The result shows that Backpropagation Neural Network can be used to classify the personality