A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information

Table 2

Experimental results.

Model	Accuracy on testing set
MNL	53.5%
NL	55.8%
FCN	57.8% (±0.33%)
RF	64.2% (±0.13%)
Customized DNN	69.5% (±0.56%)