Optimization Learning Vector Quantization Using Genetic Algorithm for Detection of Diabetics

Inggih Permana (UIN SUSKA Riau)
Nesdi Evrilyan Rozanda (UIN SUSKA Riau)
Fadhilah Syafria (UIN SUSKA Riau)
Febi Nur Salisah (UIN SUSKA Riau)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 12, No 3: December 2018

This study proposed the method to improve the result of Learning Vector Quantization (LVQ) by optimizing the weight vectors using a genetic algorithm (GA) to detect the diabetics. Initial value of individuals for GA is taken from weight vectors which come from the last m iterations of LVQ training result. The result of experiment showed that there is a significant increase in sensitivity level, however there is a significant decrease in specificity level. It means the proposed method success in improving the LVQ ability to recognized the diabetics, but it lowers the ability of LVQ to recognize the people unaffected by diabetes.This study proposed the method to improve the result of Learning Vector Quantization (LVQ) by optimizing the weight vectors using a genetic algorithm (GA) to detect the diabetics. Initial value of individuals for GA is taken from weight vectors which come from the last m iterations of LVQ training result. The result of experiment showed that there is a significant increase in sensitivity level, however there is a significant decrease in specificity level. It means the proposed method success in improving the LVQ ability to recognized the diabetics, but it lowers the ability of LVQ to recognize the people unaffected by diabetes.