Research Article
Quantitative Detection of Financial Fraud Based on Deep Learning with Combination of E-Commerce Big Data
Table 3
Comparison of actual fraud rate and predicted fraud rate in the test set.
| Sample number | Number of samples | Contains the number of frauds | Actual fraud rate | Predicted fraud rate |
| 1 | 300 | 116 | 37.632% | 35.946% | 2 | 300 | 47 | 14.246% | 14.268% | 3 | 300 | 45 | 13.596% | 12.587% | 4 | 300 | 35 | 10.175% | 10.158% | 5 | 300 | 30 | 8.457% | 8.328% | 6 | 300 | 20 | 5.082% | 4.389% | 7 | 300 | 23 | 6.125% | 5.821% | 8 | 300 | 14 | 3.059% | 2.985% | 9 | 300 | 12 | 2.379% | 3.109% | 10 | 300 | 10 | 1.695% | 1.479% |
|
|