Klasifikasi Tingkat Dehidrasi Berdasarkan Warna Urin Menggunakan Metode KNN

Abstract

The human body really needs fluids, to assist in carrying out its normal functions, if humans lack fluids, the body will experience disturbances in carrying out their functions, this fluid deficiency is accompanied by electrolyte disturbances in the body due to lack of sodium and water in the body. This condition is called dehydration. Dehydration is divided into two levels, namely mild and severe dehydration. Causes of dehydration in addition to lack of drinking fluids, dehydration can also be caused by climatic factors or hot weather and can also be caused by strenuous physical activity factors that can make humans lack body fluids through sweat. To find out whether a person is dehydrated or not, it can be seen through the color of his urine. To make it easier to classify a person is dehydrated or not, an image processing system can be used. In this study, we created a system that can classify whether a person is dehydrated or not through urine color that has been converted into a digital image and processed into a system created using Mathlab, and using the K-Nearest Neighbor method. From the results of the research that we have done, the percentage of accuracy of the K-Nearest Neighbor method is 86,67 percent with the number of accurate classifications amounting to 26 and the number of inaccurate classifications amounting to 4 out of 30 samples obtained from participants who came from college students of the STT Indonesia Tanjungpinang campus.