Jurnal Eurekamatika
Vol 12, No 1 (2024): Jurnal EurekaMatika

Implementasi CART-Real Adaboost dalam Memprediksi Minat Pelanggan Membeli Sepatu

Aprihartha, Moch. Anjas (Unknown)
Prasetya, Jus (Unknown)
Fallo, Sefri Imanuel (Unknown)



Article Info

Publish Date
02 May 2024

Abstract

Machine learning is a field of science related to the development of computer algorithms to transform data into intelligent actions. In machine learning does not escape understanding machine learning algorithms. One popular machine learning algorithm is supervised learning. Supervised learning algorithms are commonly used in solving prediction problems. This study aims to implement supervided learning algorithms using CART and CART-Real Adaboost methods in predicting customer interest in buying shoes. The results of the study obtained the performance of the CART model resulted in an accuracy of 77.5% and an AUC of 0.711 which indicates that the model is quite good. While the performance of the CART-Real Adaboost model obtained the best model at tree depth level 6 or level 8. The model obtained an accuracy of 85.71% and an AUC of 0.8225 which indicates a good model. This makes CART-Real Adaboost the best model compared to the CART model.Keywords: CART, Prediction, Real Adaboost, Shoes, Supervised Learning.AbstrakPembelajaran mesin merupakan bidang ilmu yang berkaitan pengembangan algoritma komputer untuk mengubah data menjadi suatu tindakan cerdas. Dalam pembelajaran mesin tidak luput dari memahami algoritma pembelajaran mesin. Salah satu algoritma pembelajaran mesin yang populer adalah supervised learning. Algoritma supervised learning umumnya digunakan dalam memecahkan masalah prediksi. Penelitian ini bertujuan untuk menerapkan algoritma supervided learning menggunakan metode CART dan CART-Real Adaboost dalam memprediksi minat pelanggan membeli sepatu. Hasil penelitian diperoleh performa model CART menghasilkan akurasi sebesar 77,5% dan AUC sebesar 0,711 yang menandakan model cukup baik. Sedangkan performa model CART-Real Adaboost diperoleh model terbaik pada kedalaman pohon berada di level 6 atau level 8. Model menghasilkan akurasi sebesar 85,71% dan AUC sebesar 0,8225 yang menandakan model baik. Ini menjadikan CART-Real Adaboost menjadi model terbaik dibandingkan model CART.

Copyrights © 2024






Journal Info

Abbrev

JEM

Publisher

Subject

Computer Science & IT Industrial & Manufacturing Engineering Mathematics

Description

Jurnal EurekaMatika (e-ISSN: 2528-4231, p-ISSN: 2776-480X) was first published annually on December 2013, and then since 2017 has been published twice a year, on May and November. JEM is a peer-reviewed Mathematics journal with its scope covers Algebra, Analysis, Statistics, and Applied Mathematics. ...