Skip to main content
Fig. 7 | Chinese Medicine

Fig. 7

From: Research of insomnia on traditional Chinese medicine diagnosis and treatment based on machine learning

Fig. 7

Confusion matrix. Process 1 is shown in ae. Process 2 is presented in f.In the confusion matrix, the vertical coordinate is the diagnosis made by doctors in the original medical records, and the horizontal coordinate represents the predicted value made by the random forest. The corresponding meanings of independent labels are shown in Tables 1 and 7. Taking the "cold and heat" confusion matrix (as depicted in a) in process 1 as an example, the cold and heat syndrome can be derived from the data of four diagnoses. The total number of medical record samples is 654, including 73 cases without cold and heat syndrome, 28 cases with cold syndrome, 419 cases with heat syndrome, and 134 cases with cold and heat complex syndrome. As shown in a, among the predicted values of the random forest model, the numbers of the cases accurately predicted by the random forest model for the above syndromes are 55, 14, 418 and 94 respectively. In general, a total of 581 cases are accurately predicted, and the prediction accuracy is 0.89. Similarly, the information of asthenia and sthenia (as depicted in b), five zang-organs combinations (as depicted in c), six fu-organs combinations (as depicted in Fig. 7(D)), pathogenic factors combinations (as depicted in e) and main prescription combinations (as depicted in f) can be derived from the the data of four diagnoses, and the numbers of the cases accurately predicted by the random forest model are 611, 576, 562, 557 and 559 respectively

Back to article page
\