Klasifikasi Status Gizi Balita Menggunakan Naïve Bayes dan K-Nearest Neighbor Berbasis Web

Abstract

One of the efforts in increasing the progress of a country is improving the quality of nutrition. Good balanced nutrition can increase the body's endurance, increase the level of intelligence and make a person develop normally. Infancy is a period of development and growth that requires more nutritional content than other age groups, making it the most vulnerable to nutritional disorders. Nutritional status under 5 years can be seen using the Anthroperti method. Calculations done manually make the process take a long time and errors often occur in entering data. Data mining can make decisions on the nutritional status of toddlers faster by looking at patterns in previous data. Naïve Bayes and K-Nearest Neigbor are one of the methods in classification. Naïve Bayes has the advantage of being able to achieve high accuracy values with minimal training data. Meanwhile, the K-Nearest Neighbor method has the advantage of being able to work optimally on data that has random errors (noise). Using 412 nutritional data for toddlers, the results obtained with Naïve Bayes getting an accuracy of 80.60%, while with K-Nearest Neighbor, an accuracy of 91.79% was obtained.

Machine Learning Method to Predict the Toddlers’ Nutritional Status

“…The kNN method only stores training data as a knowledge base, so it does not require a complex learning process. The kNN method can work optimally for data with noise [5].…”

“…Several ML methods are used to classify/predict malnutrition or nutritional status in toddlers, including the naïve Bayes (NB) method [3][4][5][6], logistic regression [7], k-nearest neighbor (kNN) [4,5,8], decision tree (DT) [6,[9][10][11], support vector machine (SVM) [12], and learning vector quantization [13]. Several studies compare several ML methods to classify malnutrition in toddlers [4,[14][15][16][17][18][19].…”

Sistem Pakar Diagnosa Kelainan Stunting Balita Menggunakan Metode KNN Berbasis Web

“…Penelitian untuk menentukan status stunting dengan menggunakan 412 data gizi balita mendapatkan hasil dengan Naïve Bayes medapatkan akurasi 80.60% sedangkan dengan K-Nearest Neighbor didapatkan akurasi 91,79% [8]. Kemudian penelitian dengan menggunakan parameter umur, tinggi badan dan berat badan.…”

Rancang Bangun Aplikasi Pendataan Status Gizi Pada Balita Berbasis Web Di Desa Rawalele

“…Gizi merupakan zat kimia yang berfungsi sebagai sumber energi, pertumbuhan dan pemeliharaaan jaringan sehingga tubuh dapat berjalan sebagaimana mestinya [2], [3]. Pemenuhan gizi merupakan hal penting, salah satunya pada anak karena berada pada masa yang rentan terhadap gangguan gizi [4]- [6]. Gangguan gizi buruk atau stunting merupakan kondisi kegagalan tumbuh dan kembang yang optimal pada balita sehingga berdampak pada kualitas dan kemampuan daya saing manusia, dikarenakan penurunan kemampuan dasar yang dibutuhkan dalam melakukan kegitan sehari-hari seperti kemampuan motoric [7] [8].…”