Depression Detection From Social Network Data Using Machine Learning Technique During Covid-19 Situation

Abstract

Coronavirus outbreaks have had a Signiant impact on human life. This thesis considers the detection of depression by applying a system that can effectively detected depressing text from various post in covid situation. The pandemic has an indirect effect on human mental health, which is irreversible but difficult to assess. One of the most frequent mental ailments is depression. More than 300 million individuals are said to be depressed over the world. This equates to about 4.4 percent of the global population. Facebook is great platform for people to express their emotions. We focused on the depressing post obtained from Facebook and how many individuals reacted to them in this article. We build the dataset for the training purpose by manually labelling 1000 Bengali and 1000 English Facebook post. In this study, we use machine learning algorithm, which has over-performing aspect on my classifications and a method weighing input properties implies on their importance.to classify our data and notice greater accuracy, we applied python and several common machine learning approaches.