Supervised Machine Learning Chatbots for Perinatal Mental Healthcare

Abstract

Perinatal mental health (PMH) problems are types of mood disorders which arise during pregnancy and within 24 months after the birth of a child, which affects pregnant women, newborns and family relationships. These problems may occur at any stage of maternal women. PMH is mainly diagnosed through behavioral observation, self-reporting, and behavioral scale testing. Chatbot is an effective technology. Through human-robot interaction, it can monitor the mental health status of perinatal women in real time while collecting user health data. The application of human-robot interaction in mental health services has attracted widespread attention. Compared with traditional methods, robot intervention in mental health care can help reduce the obstacles for subjects to seek help for mental health, and can collect more comprehensive and detailed data of patients, which helps users recognize their own mental health level, and can also help clinicians make diagnoses more accurately and in a timely manner. In this article, the author proposes a chatbot to monitor and assess the mental state of perinatal women. This article uses supervised machine learning to analyze the 31 characteristics of 223 samples, and trains a model to determine the anxiety, depression and hypomania index of perinatal women. Meanwhile, psychological test scales are used to assist in evaluation and make treatment suggestions to help users improve their mental health.