Selection of Home Wifi Internet: Machine Learning Implementation With Decision Tree C4.5 Algorithm Method

Abstract

The multiple bandwidths that internet service providers offer make it difficult for people to choose, especially for regular people unfamiliar with the internet; therefore, most people choose because the price is reasonable. Numerous users also lament the difficulty and slow internet usage. The issue is then concentrated on internet service providers, who are thought to be poor at offering services. The quantity of bandwidth consumed, which does not correspond to the user’s needs, is one factor contributing to slow internet. As a result, the appropriate bandwidth must be chosen based on the requirements of each user. Compared to other algorithms, the C4.5 decision tree method can deliver the best and correct decision, according to the current literature. As a result, this project will develop a web application based on the C4.5 decision tree algorithm that can assist in determining bandwidth and internet following community needs. Using this C4.5 Decision Tree, decisions are based on patterns identified in previously collected data. Predictions about various forms of internet use in the neighborhood may subsequently be produced from these patterns. Based on the calculation, the accuracy obtained is 0.54, or a percentage of 54%. The black box testing indicated that the bandwidth determination application was functioning correctly