Klasifikasi Kondisi Gizi Bayi Bawah Lima Tahun Pada Posyandu Melati Dengan Menggunakan Algoritma Decision Tree

Abstract

One of the health problems in Cirebon is about nutritional status. This happens because the increase and decrease in the number of children under five who experience nutritional status problems each year is uncertain. Toddlers are a group of people who are vulnerable to nutrition. The incidence of malnutrition if not addressed will cause a bad impact for toddlers. The impacts include death and chronic infection. Early detection of undernourished children (malnutrition and malnutrition) can be done with an examination of weight for age (W/U) to monitor the child's weight, the parameters used to calculate the nutritional status of toddlers include age, weight and height/length. To classify the nutritional status of children under five, a knowledge or scientific study is needed that can classify data based on the data from measurements and weighing. This study uses 8 criteria, namely Name, Address, Mother's Name, Gender, Age, Weight, Height, Status Classification. The accuracy results obtained are 98.86% with details, namely the Prediction Results of Malnutrition and it turns out that the True Malnutrition is 13 data. Poor Nutrition Prediction Results and turns out to be True Normal by 1 Data. Normal Prediction Results and it turns out to be True Malnutrition is 1 Data. Normal Prediction Results and turns out to be True Normal of 161 data. The results of the classification of infant levels based on age, infants aged 0 months to 10 months had normal nutrition, while infants aged 10 months to 19.5 months were prone to malnutrition for infants, and those aged more than 19.5 months had poor nutrition. normal.

Klasifikasi dan Monitoring Status Gizi Balita Melalui Penerapan Metode Naïve Bayes Classification Berbasis GIS

“…Penelitian ini akan menggunakan algoritma Naïve Bayes Classification dalam melakukan classification karena tingkat akurasi dan MAPE yang dihasilkan oleh Naïve Bayes Classifier umumnya lebih tinggi bila dibandingkan dengan KNN [4], selain itu juga Naïve Bayes Classifier memiliki performa recall terbaik bila dibandingkan dengan Decision Tree dan KNN [5]. Selain menggunakan Naïve Bayes Classification, penentuan status gizi balita juga pernah dilakukan dengan menggunakan algoritma decision tree dengan kriteria berat badan menurut umur (BB/U) dengan akurasi 98,86% [6], KNN melalui feature selection backward elimination dengan atribut identitas balita, BB/TB, z-score dari masing-masing nilai indeks Antropometrinya dan indeksasi binary stunting dan tidak stunting dengan akurasi yang dihasilkan sebesar 92,20% [7], algoritma C4.5 melalui pendekatan indeks BB/TB dengan akurasi sebesar 90% [8], serta algoritma SVM dengan kriteria yang digunakan adalah identitas balita, data posyandu, dan indeks Antropometri dengan akurasi yang dihasilkan sebesar 87% [9].…”