Anomaly Detection In Blockchain
Shafiq, Omer (2019)
2019
Tietojenkäsittelytieteiden tutkinto-ohjelma
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2019-12-13
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-201912056592
https://urn.fi/URN:NBN:fi:tuni-201912056592
Tiivistelmä
Anomaly detection has been a well-studied area for a long time. Its applications in the financial sector have aided in identifying suspicious activities of hackers. However, with the advancements in the financial domain such as blockchain and artificial intelligence, it is more challenging to deceive financial systems. Despite these technological advancements many fraudulent cases have still emerged.
Many artificial intelligence techniques have been proposed to deal with the anomaly detection problem; some results appear to be considerably assuring, but there is no explicit superior solution. This thesis leaps to bridge the gap between artificial intelligence and blockchain by pursuing various anomaly detection techniques on transactional network data of a public financial blockchain named 'Bitcoin'.
This thesis also presents an overview of the blockchain technology and its application in the financial sector in light of anomaly detection. Furthermore, it extracts the transactional data of bitcoin blockchain and analyses for malicious transactions using unsupervised machine learning techniques. A range of algorithms such as isolation forest, histogram based outlier detection (HBOS), cluster based local outlier factor (CBLOF), principal component analysis (PCA), K-means, deep autoencoder networks and ensemble method are evaluated and compared.
Many artificial intelligence techniques have been proposed to deal with the anomaly detection problem; some results appear to be considerably assuring, but there is no explicit superior solution. This thesis leaps to bridge the gap between artificial intelligence and blockchain by pursuing various anomaly detection techniques on transactional network data of a public financial blockchain named 'Bitcoin'.
This thesis also presents an overview of the blockchain technology and its application in the financial sector in light of anomaly detection. Furthermore, it extracts the transactional data of bitcoin blockchain and analyses for malicious transactions using unsupervised machine learning techniques. A range of algorithms such as isolation forest, histogram based outlier detection (HBOS), cluster based local outlier factor (CBLOF), principal component analysis (PCA), K-means, deep autoencoder networks and ensemble method are evaluated and compared.