Table 3.
Comparative analysis of the proposed method and previously published methods.
| # | Database | Number of data | Methods | Evaluation |
|---|---|---|---|---|
| [27] |
Various types of humans coronaviruses (Alpha CoV, BetaCov-1, MERS-CoV, NL63- CoV, HKU1-CoV and SARS-COV-2) |
592 (Multi-Class) |
CpG island feature selection + KNN classifier | 98 % |
| [28] |
complete genomes of COVID-19, SARS-CoV and MERS-CoV sequences |
76 | Combinatorial of DFT, DCT, and Moment Invariants techniques + KNN classifier | 100 % |
| [18] | COVID-19 and three types of Influenza viruses | 594 |
cockroach optimized deep neural network |
99 % |
| [25] | DNA sequences from 24 virus families and SARS-CoV |
347,363 (Multi-Class) |
Pseudo-convolutional method + Random Forest and MLP classifier | 99 % |
| [23] |
SARS-CoV-2, MERS-CoV, HCoV-NL63, HCoV-OC43, HCoV-229E, HCoV-HKU1, and SARS-CoV full genome |
553 (Multi-Class) |
CNN Deep learning | 98 % |
| Proposed method | coronavirus and influenza virus sequences | 107,000 | Sliding window technique on LPC model + SVM classifier | 99 % |