Research Article
A Genetic Algorithm Based Multilevel Association Rules Mining for Big Datasets
Table 4
Average fitness of association rules mined from 1000 and 2000 transactions in two datasets.
| Results from dataset 1 | Number of transactions
| Type | Time (s) | 20 | 62.5 | 125 | 250 | 400 | 1000 | 2000 | 2500 |
| 1000
| GA | 0.14 | 0.17 | 0.17 | 0.16 | 0.16 | 20 | 62.5 | 125 | AP | 0 | 0 | 0 | 0 | 0.13 | 0.15 | 0.16 | 0.16 | 2000
| GA | 0 | 0.21 | 0.24 | 0.23 | 0.22 | 0.20 | 0.20 | 0.20 | AP | 0 | 0 | 0 | 0 | 0 | 0.16 | 0.20 | 0.20 |
| Results from dataset 2 | Number of transactions
| Type | Time (s) | 15 | 20 | 40 | 100 | 200 | 400 | 600 | 1000 |
| 500
| GA | 0.35 | 0.36 | 0.35 | 0.35 | 0.35 | 0.35 | 0.35 | 0.35 | AP | 0 | 0 | 0 | 0.29 | 0.36 | 0.35 | 0.35 | 0.35 | 1000
| GA | 0 | 0.36 | 0.37 | 0.35 | 0.35 | 0.35 | 0.35 | 0.35 | AP | 0 | 0 | 0 | 0.26 | 0.28 | 0.36 | 0.35 | 0.35 |
|
|