Financial fraud prediction mechanism using machine learning and distributed processing

Dataset:
http://www.kaggle.com/ntnu-testimon/paysim1

Predicted attribute:
Class of fraud detection

Number of Instances:
6.353.307

Number of Attributes:
11 attributes (5 Decimal, 3 Integer, 3 String)

Class Distribution:
6.354.407 without fraud and 8.213 with fraud

Requirements:
Python Version: 3.6.9
Apache Spark Version: 2.4.5

---

Daniel Gleison M. Lira
Mestrado Acadêmico em Ciência da Computação
Universidade Estadual do Ceará (UECE)
[email protected]