Model structure of CNN_L12_narrow.

This table outlines the structure of CNN_L12_narrow, a variant with an additional convolutional layer compared to CNN_L11_narrow. This model also uses a learning rate of 0.0001, aiming to further refine feature extraction for a larger dataset through increased depth, potentially capturing more complex hierarchical features. #features stands for the number of input features. For instance, if the embedding method used is One-Hot, the #feature=21. (XLSX)本表格阐述了CNN_L12_narrow的网络结构，相较于CNN_L11_narrow，该变体多增加了一个卷积层。该模型同时采用0.0001的学习率，旨在通过增加网络深度，针对更大规模的数据集进一步优化特征提取流程，从而有望捕捉到更为复杂的层级特征。#features代表输入特征的数量。例如，若采用的嵌入方法为独热编码（One-Hot），则#feature=21。（XLSX）