In the following sections, we first illustrate PWCC in detail, and then we will introduce how to bag the three classifiers.
When using the PWCC model, we set the widths of three convolutional filters as 1, 2, and 3.
(iii) PWCC: we combine each review with the product to make a product word composition and then build a CNN classifier based on the composition for fake review prediction.
First, f, p, and r performance of the proposed bagging method outperforms the other methods from [BIGRAMS.sub.SVM] to PWCC. This demonstrates the effectiveness of the proposed method.
Third, the performance of PWCC performs better than both [BIGRAMS.sub.svm] and [TRIGRAMS.sub.svm].
Since the product word composition is composed of product and word information, we remove the representations P from PWCC model to build a CNN classifier based on word representation and then conduct experiments on Amazon dataset.
As shown in Figure 4, we can see that PWCC achieves better results of f, p, and r.