Implementasi Ensemble Learning untuk Mendeteksi Malware pada Android Mobile dengan Decision Tree, Naive Bayes dan Logistic Regression

Nadaputri, Gracia Angelica (2022) Implementasi Ensemble Learning untuk Mendeteksi Malware pada Android Mobile dengan Decision Tree, Naive Bayes dan Logistic Regression. Bachelor Thesis thesis, Universitas Multimedia Nusantara.

	Text HALAMAN_AWAL.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text DAFTAR_PUSTAKA.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (765kB)
	Text BAB_I.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (325kB)
	Text BAB_II.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text BAB_V.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (440kB)
	Text BAB_III.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text BAB_IV.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (5MB)
	Text LAMPIRAN.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (407kB)

Abstract

Mobile devices have been an essential part of society and with the ever-growing market share, the potential for harm has also increased. With that being said, the implementation of machine learning to aid in preventing these threats is also on the rise. One approach is to use an ensemble learning method in which smaller and lighter models, also known as base learners, are put together to predict the same problem with higher accuracy. The models used in this exploration include Decision Trees, Naive Bayes Classifiers, and Logistic Regression. Two ensemble methods were tested when put together or put in groups: Bagging and Stacking. Multiple combinations were tested with these models and methods on a dataset which consists of 50,000 data of applications, each with 1436 features. The most successful model, a Bagged Decision Tree Ensemble, produced an accuracy of 90.69%, AUC Score of 84.44%, and precision value of 91.09%.

Item Type:	Thesis (Bachelor Thesis)
Keywords:	Android Permission-Based Malware Detection, Decision Trees, Ensemble Learning, Logistic Regression, Naive Bayes Classifier
Subjects:	000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 005 Computer Programming 000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 005 Computer Programming > 005.5 Application / Software
Divisions:	Faculty of Engineering & Informatics > Informatics
SWORD Depositor:	Administrator UMN Library
Depositing User:	Administrator UMN Library
Date Deposited:	19 Mar 2024 07:09
Last Modified:	19 Mar 2024 07:14
URI:	https://kc.umn.ac.id/id/eprint/28069

Actions (login required)

View Item