A novel automated depression detection technique using text transcript

摘要

Depression is one of the most common mental illnesses, impacting billions ofpeople worldwide. The lack of existing resources is impeding the country's economicprosperity. As a result, new approaches for detecting and treating mentaldiseases as well as reaching out to individuals are required so that peoplecan overcome their daily challenges and become more productive. An automateddepression detection system can greatly aid in clinical findings and earlytreatment of depression. Automatic detection, like in a clinical interview canbe derived from various modalities that include video, audio, and text. Amongthese modalities, audio characteristics are the most commonly researchedwhile text elements are seldom investigated. In the light of the above, thispaper proposes a novel automated depression identification approach based onlinguistic material gathered from patient interviews. The focus is to enhanceboth the accuracy and efficiency of detection. The proposed approach is madeup of two parts: a Bidirectional Gated Recurrent Unit (BGRU) network fordealing with linguistic information and a fully coupled network that integratesthe model outputs to measure the depressed state. The proposed approach isvalidated using Distress Analysis Interview Corpus-Wizard-of-Oz interviewsdataset. To evaluate the performance precision, recall, and F1 score are computedusing the proposed approach and then the comparative analysis is donewith the existing approaches. The proposed approach yielded an F1 score of0.92, indicating the existence of depression as well as the projected severitylevel. It is realized from the generated results that the proposed approach hasoutperformed the previous ones. The proposed approach can not only automaticallyassess the severity of depression but also enhances both the accuracyand efficiency of detection. The proposed approach indicates the feasibility ofBGRU over Long Short Term Memory by achieving exceptional results for recognitionof depression.