Fig 1.
Ecological discourse analysis framework of attitude resources in news discourse.
Fig 2.
The constructed neural network model.
Table 1.
Parameter settings of CNN.
Fig 3.
Change and distribution of nonlinear excitation function based on CNN.
a. Learning rate is 0.001; b. Learning rate is 0.01.
Fig 4.
The trend of accuracy and loss rate of Chinese and English texts.
a. Accuracy (Chinese); b. Loss rate (Chinese); c. Accuracy (English); d. Loss rate (English).
Table 2.
Comparison of performance of the models.
Table 3.
Ranking of beneficial ecological discourse-emotional words.
Table 4.
Arrangement of neutral ecological discourse-emotional words.
Table 5.
Ecological orientation of attitudinal resource refraction in news discourses of the Chinese climate change.