Sensitivity-based data selection for predicting individual´s sub-health on TCM doctors´ diagnosis

In this paper we propose an approach of predicting individual´s sub-health based on the principle of TCM (Traditional Chinese medicine) as a preventive medicine. The object´s vision features like features of tongue, eye and face are extracted for modeling a process of TCM doctor´s diagnosis. As a consequence of the diversity and uncertainty of TCM doctors´ diagnosis, the sensitivity is defined as a criterion to select the appropriate features from the derived features, and integrate the diagnosis data given by multiple doctors as training data for constructing the sub-health inference model. The experiment results show that the sensitivity-based feature selection and diagnosis data integration improve the model´s inference performance on the accuracy, correlation and residual variance.