Analisis Malware pada Perangkat Internet of Things Berbasis Cascade dengan Model RNN-LSTM

Ilham, Karina Fitriwulandari (2025) Analisis Malware pada Perangkat Internet of Things Berbasis Cascade dengan Model RNN-LSTM. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6025231056-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy

Abstract

Dalam era digital yang terus berkembang, Internet of Things (IoT) telah menjadi pilar utama transformasi teknologi dengan berbagai aplikasi yang mencakup sektor kesehatan, transportasi, hingga pemerintahan. IoT memungkinkan perangkat berkomunikasi dan bertukar informasi melalui jaringan internet, namun ketergantungan ini juga menghadirkan risiko keamanan yang signifikan. Salah satu ancaman utama adalah malware yang menargetkan perangkat IoT untuk melakukan tindakan ilegal seperti pencurian data atau serangan berbahaya lainnya. Interkonektivitas perangkat IoT menciptakan banyak celah yang dapat dimanfaatkan oleh penyerang, mengingat sebagian besar perangkat ini menyimpan data pribadi yang sensitif. Dengan variasi jenis malware yang beragam serta kompleksitas dan volume data yang besar, penelitian ini mengembangkan pendekatan cascade berbasis RNN-LSTM. Pendekatan ini dirancang untuk mendeteksi dan mengklasifikasikan berbagai jenis malware. Model ini tidak hanya memisahkan lalu lintas jaringan benign dan malware, tetapi juga mampu mengenali kelas malware menyerang perangkat IoT. Hasil penelitian menunjukkan bahwa pendekatan cascade berbasis RNN-LSTM mencatat rata-rata F2-Score sebesar 79,95%. Selain itu, model cascade juga mencatat Precision, Recall, F1-Score, dan akurasi rata-rata masing-masing sebesar 80,15%, 81,75%, 78,29%, dan 81,79%. Walaupun pendekatan cascade memerlukan waktu eksekusi yang lebih lama dan konsumsi memori yang lebih tinggi.
======================================================================================================================================
In the continually evolving digital era, the Internet of Things (IoT) has become a main pillar of technological transformation with applications spanning health care, transportation, and government sectors. IoT allows devices to communicate and exchange information over the internet; however, this dependency also introduces significant security risks. One of the primary threats is malware targeting IoT devices to perform illegal actions such as data theft or other malicious attacks. The interconnectivity of IoT devices creates numerous vulnerabilities that attackers can exploit, especially since many of these devices store sensitive personal data. Given the diverse types of malwares, as well as the complexity and volume of data, this research develops a cascade approach based on RNN-LSTM. This approach is designed to detect and classify various types of malwares. The model not only separates benign and malware network traffic but also identifies the classes of malware attacking IoT devices. The study results show that the RNN-LSTM-based cascade approach achieves an average F2-Score of 79.95%. Additionally, the cascade model records average Precision, Recall, F1-Score, and accuracy rates of 80.15%, 81.75%, 78.29%, and 81.79%, respectively, albeit with longer execution times and higher memory consumption.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Cascade, Dataset IoT-23, Internet of Things (IoT), Keamanan IoT, Klasifikasi, RNN-LSTM. ======================================================================================================================= Cascade, Classification, IoT-23 Dataset, Internet of Things (IoT), IoT Security, RNN-LSTM.
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76.9.A25 Computer security. Digital forensic. Data encryption (Computer science) T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55101-(S2) Master Thesis
Depositing User:	Karina Fitriwulandari Ilham
Date Deposited:	28 Jan 2025 02:42
Last Modified:	28 Jan 2025 02:42
URI:	http://repository.its.ac.id/id/eprint/117002

Actions (login required)

View Item