Table 4 Reduction (%) of rows, and storage bytes. Comparison between the original IoT-23 dataset and the generated CGNs
From: A novel botnet attack detection for IoT networks based on communication graphs
Dataset | Data flows | CGN | % Reduction | |||
|---|---|---|---|---|---|---|
Rows | Bytes | Rows | Bytes | Rows | Bytes | |
44-01_IoTMalware | 211 | 24,978 | 9 | 616 | 95.73% | 97.53% |
04-01_Honeypot | 452 | 57,855 | 454 | 32,349 | − 0.44% | 44.09% |
05-01_Honeypot | 1374 | 170,507 | 944 | 66,922 | 31.30% | 60.75% |
20-01_IoTMalware | 3193 | 388,802 | 628 | 45,833 | 80.33% | 88.21% |
21-01_IoTMalware | 3272 | 398,067 | 1924 | 142,330 | 41.20% | 64.24% |
42-01_IoTMalware | 4420 | 546,951 | 3099 | 226,308 | 29.89% | 58.62% |
08-01_IoTMalware | 2181 | 271,260 | 10 | 824 | 99.54% | 99.70% |
34-01_IoTMalware | 1923 | 223,957 | 293 | 20,869 | 84.76% | 90.68% |
03-01_IoTMalware | 4536 | 534,951 | 5609 | 390,963 | − 23.66% | 26.92% |
01-01_IoTMalware | 469,275 | 53,974,498 | 441,334 | 32,239,514 | 5.95% | 40.27% |
60-01_IoTMalware | 2476 | 284,405 | 63 | 4579 | 97.46% | 98.39% |
48-01_IoTMalware | 3734 | 445,406 | 2319 | 159,938 | 37.90% | 64.09% |
49-01_IoTMalware | 3665 | 443,332 | 3116 | 222,321 | 14.98% | 49.85% |
09-01_IoTMalware | 22,548 | 2,787,317 | 18,639 | 1,333,986 | 17.34% | 52.14% |
35-01_IoTMalware | 8,262,389 | 958,410,934 | 4,120,109 | 456,537,432 | 50.13% | 52.37% |
07-01_IoTMalware | 75,955 | 9,676,495 | 113,538 | 7,911,788 | − 49.48% | 18.24% |
36-01_IoTMalware | 2663 | 306,067 | 1170 | 81,773 | 56.06% | 73.28% |
52-01_IoTMalware | 1794 | 210,160 | 1369 | 94,072 | 23.69% | 55.24% |
33-01_IoTMalware | 1,380,791 | 154,919,240 | 1,362,849 | 97,414,905 | 1.30% | 37.12% |
17-01_IoTMalware | 31,438 | 3,933,107 | 25,206 | 1,834,122 | 19.82% | 53.37% |
39-01_IoTMalware | 7337 | 870,739 | 8,534 | 613,082 | − 16.72% | 29.59% |