Table 1.
Spammer behavior used in literature.
Table 2.
Amazon dataset used in the proposed framework.
Fig 1.
SD-FSL-CLSTM spam detection framework.
Table 3.
Derived feature.
Table 4.
Notation used in methodology.
Fig 2.
Evaluation of individual features using CRF.
Table 5.
Feature selection using XGB.
Table 6.
Feature selection using PCA.
Table 7.
Feature scoring using XGB.
Table 8.
Feature scoring using PCA.
Fig 3.
Feature scoring using CRF PCA, XGB.
Table 9.
Evaluation matrices of the proposed approach.
Table 10.
Results on evaluation metrics of proposed deep learning methods on linguistic features.
Table 11.
The proposed approach and state-of-the-art study results comparison.