基于BiLSTM和CNN的序贯三支情感分类模型研究

赵梦宇, 孙京博, 魏遵天, 辛现伟, 宋继华

Research on sequential three⁃way sentiment classification model based on BiLSTM and CNN

Mengyu Zhao, Jingbo Sun, Zuntian Wei, Xianwei Xin, Jihua Song

表3 三个数据集在不同模型下分类召回率对比

Table 3 Cassification recall of three datasets with different models

模型 Hotel Computer 中文微博情感 BiLCNN⁃S3WD 0.9165±0.0023 0.9280±0.0137 0.8190±0.0059 BiLSTM 0.9114±0.0037 0.9085±0.0101 0.7959±0.0102 BiLSTM+CNN (kernel_size=2) 0.9130±0.0033 0.9194±0.0015 0.7591±0.0226 BiLSTM+CNN (kernel_size=3) 0.9125±0.0034 0.9216±0.0055 0.7997±0.0134 BiLSTM+CNN (kernel_size=4) 0.9145±0.0027 0.9209±0.0040 0.7946±0.0175 BiLSTM+CNN (kernel_size=5) 0.9140±0.0030 0.9194±0.0073 0.8085±0.0118 DLSTWSC 0.9110±0.0020 0.9259±0.0000 0.7806±0.0250 CFRT 0.7377±0.0039 0.7581±0.0046 0.6564±0.0021