Ανάλυση συναλλαγών Ethereum για εντοπισμό ανωμαλιών

Κοντός, Παναγιώτης

Analyzing Ethereum transactions for anomaly detection

Master Thesis

Author

Κοντός, Παναγιώτης

Date

2025-06

Abstract

This document delves into the analysis of Ethereum blockchain transactions to uncover‬ ‭ anomalies using advanced machine learning techniques. The aim is to identify unusual patterns‬ ‭ that might signal fraudulent activities or irregularities within the transactional data. Through‬ ‭ detailed feature engineering and exploratory data analysis (EDA), the study uncovers critical‬ ‭ insights into transaction behaviors. Three algorithms DBSCAN, Isolation Forest, and K-Nearest‬ ‭ Neighbors (KNN) were evaluated for anomaly detection. Among them, KNN emerged as the‬ ‭ most reliable, offering robust accuracy and recall metrics. These findings pave the way for‬ ‭ enhanced blockchain security and more trustworthy transaction networks.‬

Postgraduate Studies Programme

Προηγμένα Συστήματα Πληροφορικής

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Πληροφορικής

Number of pages

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/17992

Collections

Τμήμα Πληροφορικής

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού 3.0 Ελλάδα