Name
Abstract | ||
|---|---|---|
Many people have dealt with fraud through their bank or credit card company, but are unaware of how it is detected by these entities. This work aims to clearly explore the possible methods used in exposing fraud and form an idea as to what qualities are attributed to it. The data sets we used had multiple useful attributes as well as plenty of entries to train our machine learning algorithms. Our research first led us to find what attributes have the highest correlation to fraud. This was found by creating a correlation chart in our descriptive analysis. That information was then used to train our machine learning algorithms during our predictive and prescriptive analyses. Decision Tree, Linear Regression, and Logistic Regression algorithms were used on the data in combination to reach a surprising level of accuracy. This line of research gives more in-depth information to the general populous who may be struggling with fraud related issues. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/CCWC51732.2021.9376045 | 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC) |
Keywords | DocType | ISBN |
Fraud detection,PaySim,Synthetic data,Financial data | Conference | 978-1-6654-3058-6 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
7 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Nicholas Cochrane | 1 | 0 | 0.34 |
| Thomas Gomez | 2 | 0 | 0.34 |
| John Warmerdam | 3 | 0 | 0.34 |
| Moises Flores | 4 | 0 | 0.34 |
| Preston Mccullough | 5 | 0 | 0.34 |
| Vincent Weinberger | 6 | 0 | 0.34 |
| Matin Pirouz | 7 | 0 | 0.34 |