Open AccessPERANCANGAN SISTEM KLASIFIKASI UDANG BERACUN PADA JENIS UDANG TENGGEK MENGGUNAKAN METODE K-NEAREST NEIGHBOR (K-NN).
Author(s) -
Lisa Afrinanda,
Ilyas Ilyas
Publication year - 2020
Publication title -
selodang mayang
Language(s) - English
Resource type - Journals
eISSN - 2620-3332
pISSN - 2442-7845
DOI - 10.47521/selodangmayang.v6i1.140
Subject(s) - shrimp , rgb color model , k nearest neighbors algorithm , biology , fishery , mathematics , pattern recognition (psychology) , artificial intelligence , computer science
Shrimp is one of the seafood which is nutrient-rich needed by the body. However, due to the frequent case of the infected Tenggek-shrimp appeared, it makes people beware to consume it. The classification of Tenggek-shrimp by using image processing of the computer be able to classify the types of shrimp whether poisonous or not. The data mining techniques can be used to classify shrimp based on RGB colors (red, green, blue) and texture (energy, contrast, correlation, homogeneity). The class of Tenggek-shrimp is divided into two, The fresh Tenggek-shrimps that are caught naturally (Class A) and the poisoned Tenggek-shrimps that are caught by using the poison (Class B). The method used in this study is K-Nearest Neighbor (K-NN). This classification system is expected to help the people in selecting good and safe Tenggek-shrimp for consumption. Based on the evaluation results using the holdout method, obtained an average accuracy of 63% with an accuracy of identification of toxic tenggek shrimp of 71.66%, and the accuracy of identification of natural fresh shrimp is about 60%.