A method for obtaining a classifier for recognizing emotions

Abstract

The present invention provides a method for obtaining a classifier for identifying emotions, including obtaining a first sample set and a second sample set; each element in the first sample set and the second sample set includes four contents, respectively is the first eigenvalue based on heart rate variability, the second eigenvalue based on the skin conductance signal, the third eigenvalue based on the skin conductance signal, and the emotion classification result; obtain the training process control parameters and initialize the loop control parameters and classifier; execute Cycle the training process to train sub-classifiers that meet the preset requirements; cascade the sub-classifiers with the current classifier to update the current classifier; update the misjudgment rate and detection rate of the current classifier; judge the current Whether the misjudgment rate of the classifier is greater than the maximum misjudgment rate; if not, output the current classifier. The invention pays attention to misjudged samples and samples expressing negative emotions, so that the classifier can absorb mistakes and improve the sensitivity of identifying negative emotions.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method for acquiring a classifier for identifying emotions. Background technique [0002] With the increasing popularity of smart wearable devices, the data processing of physiological parameters has increasingly become a research hotspot. Data processing based on physiological parameters can facilitate the grasp of the user's physical state and psychological state, thereby providing users with various information based on Physical state and psychological state services, for example, if the user is in a bad mood, it can play cheerful songs for the user to adjust the body and mind, and can also conduct intelligent dialogues with the user according to the user's physical state and psychological state. Therefore, the data processing of physiological parameters has a relatively strong High research value; [0003] However, the relevant technologies for data processing of physiologic...

AI Technical Summary

Problems solved by technology

[0003] However, the relevant technologies for data processing of physiological parameters in the existing technology are not mature, which limits the research and development of technical solutions that provide related services based on the user's physical state and psychological state.

Images