Comparing Supervised Classification Algorithms in Machine Learning for Poverty Prediction

Abstract

Due to their capacity to evaluate enormous datasets and generate precise predictions, machine learning algorithms have attracted a lot of interest lately. These algorithms have been used in a variety of fields, including social sciences, finance, healthcare, and marketing. Machine learning algorithms offer a viable method for dividing families into poor and non-poor groups based on pertinent socioeconomic characteristics in the context of poverty studies. This research assesses the performance of various surprised classification algorithms machine learning peculiarly Naïve Bayesian Algorithms, Support Vector Machines, K Nearest Neighbor, Decision Trees, and Logistic Regression and Bagging algorithms in predicting poverty degree. Empirical findings demonstrate that the model with the highest accuracy is Decision Tree, with an accuracy of 0.9961. This means that 99.61% of the instances were correctly classified by Decision Tree. The model with the lowest accuracy is Naive Bayes, with an accuracy of 0.5103. This means that only 51.03% of the instances were correctly classified by Naive Bayes.