Research Article

Knowledge-Based Recurrent Neural Network for TCM Cerebral Palsy Diagnosis

Table 1

The classification accuracy of the KBRNN and baselines.

PresetPreset + 50% train-setPreset + 100% train-set

KBRNN79.3182.0383.12
RNN44.2871.3278.03
LSTM45.0176.0479.49
GRU45.1976.2279.85
Bi-RNN44.6476.5982.39
Bi-LSTM45.7477.1380.40
Bi-GRU45.9179.8582.58
CNN47.3779.6781.67
DAN48.0980.2181.85