DCNN (Deep Convolutional Neural Network)-DNN (Deep Neural Network) and PV-SVM (Paragraph Vector-Support Vector Machine)-based multi-modal depressive disorder estimation and classification method

Abstract

The invention relates to a DCNN (Deep Convolutional Neural Network)-DNN (Deep Neural Network) and PV-SVM (Paragraph Vector-Support Vector Machine)-based multi-modal depressive disorder estimation and classification method. The method comprises the following steps: preprocessing audio and video features through a displacement range histogram and an Opensmile tool, extracting hidden layer abstract features of audio and video statistical features through a DCNN, performing depressive disorder estimation through a DNN, performing high-dimensional feature mapping on textile information through a PV method, inputting an obtained high-dimensional feature expression into an SVM for binary classification, connecting a depressive disorder estimation result with a binary classification result in series, inputting the whole into a random forests model for training, and performing a final depressive disorder classification task through the trained random forests model, namely judging a depressive disorder or a non depressive disorder. By the adoption of a DCNN model for extraction of the hidden layer abstract features from a primary audio / video, an original high-dimensional feature is more compact, and included information is richer; therefore, the model is more effective, and the phenomenon of overfitting caused by extremely high dimension of the feature is avoided.

Description

technical field [0001] The invention belongs to the combined field of computer and medicine, adopts deep convolutional neural network (DCNN), deep neural network (DNN), paragraph vector (PV) and support vector machine (SVM) model, and relates to a methods for classifying depression. Background technique [0002] In recent years, the field of artificial intelligence has used machine learning methods to establish a variety of depression detection systems starting from audio and video to help psychologists detect, prevent and treat clinical depression. In the past few years, many important results have been achieved in the field of depression detection through audio-visual information. The document "DecisionTree Based Depression Classification from Audio Video and Language Information, 2016 6th AVEC, pp 89-96" discloses a multimodal depression estimation based on audio and video, and then manually builds a decision tree to classify depression based on text information . This...

AI Technical Summary

Problems solved by technology

This many-to-one problem causes a lot of useful timing information to be discarded, resulting in information waste

In order to make up for the shortcoming of information loss, scholars have designed a variety of statistical methods, which produce very high feature dimensions and easily lead to overfitting.

[0005] 2) A large amount of useful information has not been mined

[0006] In summary, the existing depression detection methods are easily affected by the feature dimension, and a large amount of information has not been fully studied, which may easily lead to low accuracy of depression estimation and low accuracy of depression detection and classification

Images