Research Article

Deep-Stacking Network Approach by Multisource Data Mining for Hazardous Risk Identification in IoT-Based Intelligent Food Management Systems

Table 3

Identification result of the risk level in comparative methods.

ModelsAccuracy (%)Model parameters (megabyte)
Architecture-1Architecture-2Architecture-1Architecture-2

Logistic regression [19]59.2169.3541.3143.25
K-nearest neighbour [35]66.3269.0150.6561.36
Support vector machine [20]66.7072.6047.3358.97
Extra trees [18]71.8872.8363.5770.65
Gradient boosting [22]78.9481.2961.3766.91
Random forest [26]80.5183.9370.1881.32
Decision tree [17]83.9785.4572.3479.25
AlexNet [26]70.6271.32218.96223.54
VGG [28]72.8873.94320.78365.12
GoogLeNet [27]85.1887.98425.89478.23
ResNet [29]87.2689.68507.34528.67
Deep forests [32]90.7392.47385.32419.11
Proposed DSN94.8897.62202.53211.26