Research Article
Beyond Words: An Intelligent Human-Machine Dialogue System with Multimodal Generation and Emotional Comprehension
Table 1
Objective experimental results of different methods (higher values indicating better models).
| | Model | Bleu-2 | Nist | Rouge-L | Meteor | Dist-2 | Ent-4 | Coherence | Fluency |
| | Text-based | 0.2921 | 2.884 | 0.0157 | 0.0098 | 0.0069 | 1.327 | 0.1906 | 0.1697 | | Emotion-based | 0.303 | 2.8012 | 0.0169 | 0.0101 | 0.0082 | 1.379 | 0.2089 | 0.1731 | | Image-Chat | 0.3313 | 3.0531 | 0.0199 | 0.0128 | 0.01 | 1.621 | 0.2311 | 0.2097 | | Ours-Emotion | 0.3055 | 2.9805 | 0.0168 | 0.0102 | 0.0077 | 1.38 | 0.1909 | 0.1718 | | Ours-Visual | 0.3243 | 3.0104 | 0.0188 | 0.0116 | 0.0093 | 1.67 | 0.2038 | 0.1879 |
| | Ours | 0.3665 | 3.2483 | 0.0225 | 0.0147 | 0.0112 | 1.86 | 0.2696 | 0.2306 |
|
|