Comparative Analysis of Machine Learning Algorithms on Family Wellness Classification

Comparative Analysis of Machine Learning Algorithms on Family Wellness Classification

Blog Article

Family welfare is Muffin a state in which a family can experience happiness, have a decent quality of life, and be sufficient in meeting primary and secondary needs in family life.One factor that influences family welfare is the amount of per capita expenditure.This study aims to compare the performance of three machine learning algorithms, namely KNN (K-Nearest Neighbors), random forest, and naive Bayes, in classifying the status of families per province in Indonesia as prosperous or not prosperous.The data used in this study is demographic and social statistics data from the years 2017-2021, obtained from the bps.go.

id website.The first statistical analysis conducted is Bloomer Set principal component analysis (PCA) with 9 predictor variables.PCA produces four principal components which are then used in the KNN, random forest, and naive Bayes methods.The analysis results from the KNN, random forest, and naive Bayes methods each yield an F1-score of 65.46%, 68%, and 69.

44%, respectively.