Table 1.
Comparative analysis of existing works with their supporting features.
Table 2.
Defining threshold range for smart bins parameters.
Table 3.
Scenarios to bin metrics mapping and labeling.
Table 4.
Poisonous gas range and threshold details.
Table 5.
Garbage waste dataset sample collected from sensors for the proposed model.
Fig 1.
Architecture of automatic waste management system.
Fig 2.
Work flow modules of the developed model.
Fig 3.
Random forest model demonstration.
Fig 4.
Bin load status after 2 hours of deployment.
Fig 5.
Bin load status after 4 hours of deployment.
Fig 6.
Bin load status after 6 hours of deployment.
Fig 7.
Time analysis of bin fill up status.
Table 6.
Bins that were never filled up after 6 hours.
Fig 8.
Classification accuracy analysis with respect to decision trees used in random forest algorithm.
Fig 9.
Execution delay analysis with respect to decision trees used in random forest algorithm.
Table 7.
Performance metrics comparison with different predictive models.
Fig 10.
Model training and testing latency analysis with different algorithms.
Fig 11.
Accuracy analysis in context with the threshold value of smart bin.