DETEKSI TRANSAKSI ANOMALI PADA BLOCKCHAIN DENGAN MENGGUNAKAN METODE LONG SHORT TERM MEMORY (LSTM)

WAHYUDI, MUHAMMAD TRI and Stiawan, Deris and Ubaya, Huda (2023) DETEKSI TRANSAKSI ANOMALI PADA BLOCKCHAIN DENGAN MENGGUNAKAN METODE LONG SHORT TERM MEMORY (LSTM). Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011381924102.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_56201_09011381924102_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_56201_09011381924102_0003047905_0216068101_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (696kB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (676kB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (185kB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (311kB) | Request a copy
	Text RAMA_56201_09011381924102_0003047905_0216068101_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (947kB) | Request a copy

Abstract

The rapid growth and anonymity offered by cryptocurrencies have made them vulnerable to being used for illegal activities. In this study, Long Short-Term Memory (LSTM) neural network is used to identify anomalous transaction patterns in the dataset. The dataset is created by retrieving raw data based on year parameters to extract a subset of the data. These subsets were extracted by writing python code snippets and represent data from 2011 to 2013. The dataset was class balanced by oversampling and undersampling techniques. The best model achieved an accuracy of 85.67%. Then, through k-fold validation, the model showed good consistency, with an average accuracy of 85.80%. These results indicate that the model has consistent and reliable performance in the given detection task.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Blockchain, Deteksi Anomali, Long Short-Term Memory
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.A25 Computer security. Systems and Data Security.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Muhammad Tri Wahyudi
Date Deposited:	22 Nov 2023 01:49
Last Modified:	22 Nov 2023 01:49
URI:	http://repository.unsri.ac.id/id/eprint/130851

Actions (login required)

View Item