Table 2 Comparison of Real and Simulated Results. The best results are highlighted by Bold and the second-best results are marked by Italic.
From: Social opinions prediction utilizes fusing dynamics equation with LLM-based agents
Method | Pangmao | Jiangping | Qingdao | Dianduji | ||||
|---|---|---|---|---|---|---|---|---|
DTW\(\downarrow\) | Corr.\(\uparrow\) | DTW\(\downarrow\) | Corr.\(\uparrow\) | DTW\(\downarrow\) | Corr.\(\uparrow\) | DTW\(\downarrow\) | Corr.\(\uparrow\) | |
FDE-LLM | 0.3622 | 0.9653 | 0.3664 | 0.8950 | 0.3352 | 0.9605 | 0.3404 | 0.8842 |
LLM | 2.6584 | 0.7139 | 2.9963 | 0.7208 | 2.7316 | 0.7511 | 2.3170 | 0.7819 |
ABM(CA) | 0.4335 | 0.8920 | 0.7648 | 0.6748 | 0.2904 | 0.6800 | 0.6806 | 0.8282 |
ABM(HK) | 1.1838 | 0.6475 | 0.6085 | 0.6961 | 1.1470 | 0.6874 | 0.9453 | 0.6760 |
ABM(PM) | 2.5653 | 0.1591 | 2.7977 | 0.3866 | 2.6302 | 0.0375 | 2.9916 | 0.2489 |
ABM(RLE) | 0.9592 | 0.9489 | 1.3978 | 0.9321 | 1.8712 | 0.9358 | 0.9180 | 0.8304 |
LLM+ABM(CA) | 0.7463 | 0.5225 | 1.5748 | 0.6425 | 0.5497 | 0.9174 | 0.8684 | 0.7386 |
