Table 1.
Quick analysis of literature survey.
| Authors | Techniques incorporated | Performance metrics analysis | Advantages | Disadvantages | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Accuracy | Specificity | Sensitivity | Precision | F1-score | Computational complexity | Time complexity | ||||
| Jenifer et al. (2022) | Edge base detection using IoT |
 High |
High |
Average |
Average |
 Low |
High |
High |
High learning rate | Not suitable for real time environment |
| Raju et al. (2022) | CCNN + GSO optimization |
Average |
Average |
Average |
Average |
Low |
Average |
High |
Supports real time environment | Prediction efficiency needs to be improvised |
| Gupta et al. (2023) | CNN |
Average |
Not analyzed |
Low |
Average |
High |
High |
Average |
Handle high scale datasets | More energy consumption |
| Sarmah (2020) | DLMNN |
Average |
Low |
Low |
Not analyzed |
Average |
High |
High |
High security | Does not incorporated any optimization which increased the training time |
| Mansour et al. (2021) | CSO-CLSTM |
Average |
Low |
Low |
Not analyzed | Not analyzed |
High |
High |
Required less energy consumption | Security of the IoT data were not considered |
| Lavanya et al. (2023) | DWT + CNN |
Average |
Average |
Low |
Not analyzed | Not analyzed |
High |
Average |
Required less time for training | The system does not have compatibility with edge devices |
| Venkatesan et al. (2022) | FNN + DCNN |
Low |
Not analyzed | Not analyzed |
low |
low |
Average |
High |
It required less attributes for training | The effectiveness of this technique in older people with persistent heart conditions has not been tested. |
| Minh et al. (2022) | Fog-based IoT approach |
Average |
Not analyzed | Not analyzed | Not analyzed | Not analyzed |
Average |
High |
Guaranteed integrity of data and support real time environment | Required more attributes to achieve high accuracy |
| Verma et al. (2022) | FETCH |
Average |
Not analyzed | Not analyzed | Not analyzed | Not analyzed |
High |
Average |
Improves the computing efficiency | High Connectivity issues |
| Nancy et al. (2022) | Bi-LSTM |
High |
Average |
Low |
Average |
Low |
High |
High |
Required less timing for training | Not suitable for real time environment |
| Srinivasu et al. (2022) | 6G Networks |
High |
High |
Low |
Average |
Average |
High |
High |
It required less attributes for training | High Connectivity issues |
| Hussien et al. (2022) | HHO |
High |
Average |
Average |
Average |
Average |
Average |
Average |
Required less timing for training. | This framework does not covered more real time frameworks. |
| Muttaqin et al. (2023) | CNN |
High |
Not analyzed | Not analyzed | Not analyzed | Not analyzed | Not analyzed | Not analyzed | It required less attributes for training | Not suitable for real time environment |
| Tair et al. (2022) | Chaotic enabled Whale optimization |
High |
Not analyzed | Not analyzed | Not analyzed | Not analyzed | Not analyzed | Not analyzed | High accuracy | Needs improvisation for IoT environment |
| Basha et al. (2021) | Chaotic HHA |
High |
Acceptable | Acceptable | Not analyzed | Not analyzed | Not analyzed | Not analyzed | Improved performance over the other algorithms | Needs improvisation for IoT environment |