Table 7 Relative change in F1 score using C+M for property crime prediction in Chicago in each test month
From: Enhancing short-term crime prediction with human mobility flows and deep learning architectures
Model | Aug | Sept | Oct | Nov | Dec |
|---|---|---|---|---|---|
HALR | 3.7% | 3.0% | 1.4% | 2.3% | 2.7% |
GRU | 4.0% | 4.3% | 2.5% | 3.9% | 2.4% |
Attn | 2.9% | 2.2% | 2.2% | 3.2% | 3.7% |
GCN | 1.2% | 1.0% | 0.3% | 0.1% | −1.8% |
GGConv | 1.2% | 0.3% | 0.9% | −0.6% | 1.3% |
NbConv | 4.3% | 4.9% | 5.6% | 4.2% | 4.5% |