Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics
Achmad Junaidi
Department of Informatics Engineering, Faculty of Computer Science, UPN “Veteran” Jawa Timur, Surabaya, Indonesia
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Health Professions Materials Science & Nanotechnology

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Bitcoin Mining Hardware Profitability Prediction Using Categorical Boosting and Extreme Gradient Boosting Algorithms Dimas Satria Prayoga; Anggraini Puspita Sari; Achmad Junaidi
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 7 No. 1 (2025): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/9xb2dz14

Abstract

Cryptocurrencies, especially Bitcoin, have gained global recognition, with mining being one of its most interesting aspects. This is especially important in the context where only a few types of bitcoin mining rigs are expected to operate profitably. On the other hand, in the field of machine learning, there are widely used algorithms, namely Extreme Gradient Boosting (XGBoost), which is known for its effectiveness, and Categorical Boosting (CatBoost), which excels in handling categorical data. This study aims to combine the performance of CatBoost and XGBoost using the Ridge Regression technique in predicting a case study that is not often encountered, namely predicting the profitability of Bitcoin mining hardware. The main steps include collecting data from reliable sources, preprocessing the data to ensure compatibility, feature selection to select the most relevant features, building a prediction model using the preprocessed data set, and then training and testing both models to evaluate their predictive accuracy. The evaluation metrics on the test data reveal the performance of CatBoost, XGBoost, and the CatBoost-XGBoost. CatBoost demonstrates a training time of 3.35 seconds with a MAPE of 15.67% and an RMSE of 0.1733. In comparison, XGBoost has a longer training time of 5.27 seconds but achieves a significantly lower MAPE of 6.49% and an RMSE of 0.1737. Meanwhile, the CatBoost-XGBoost, with the longest training time of 6.84 seconds, delivers a competitive MAPE of 6.57% and the lowest RMSE of 0.1696 among the three approaches. These results highlight that while XGBoost and CatBoost meta model outperform CatBoost in terms of accuracy, the Ridge meta model provides slightly better overall predictive performance based on RMSE.
Co-Authors Anggraini Puspita Sari Dimas Satria Prayoga