Depression Detection in Social Media Comments Data Using Machine Learning Algorithms

Abstract

Nowadays, depression is a common and dangerous mental problem for our society, the country even the whole world. When a person is in a heartbreaking mood or going through an exquisite condition and it is not leaving him, trying to live alone, and giving him pain continuously is called depression. The last stage of depression is killing himself. According to WHO, currently,4.4 of people worldwide suffer from depression. Many depressed people die almost every day. So, we will generate a model to find out who is suffering from depression and who is not. And finding depression is quite easy through our model. We collected huge data from Facebook, YouTube, and social media for the buildup models and learn to model and machine. Here we applied six classifiers to detect depression such as SVM, DT, LR, KNN etc. And when we are searching for which classifier gives the best accuracy then we see that the Support Vector Machine gives the best accuracy and which is 75%.