Predicting Depression in Social Network Sites Using NLP

Abstract

Depression is an acute problem throughout the world, where more than 264 million people are suffering from it. Due to worst and prolong depression near around 800000 people dies in every year. The real problem is that most of the people are not concern of the fact that they are suffering from depression. Here, our aim was to find out whether an individual is in depression or not by analyzing social media text information. Our dataset consists of 1500 sentences, which was collected from different social media platforms– Facebook, Tweeter, and Instagram. Then we have performed some data preprocessing approaches such as– tokenization, remove of stop words, remove of empty string, remove of punctuations, stemming and lemmatizing. After data preprocessing, we considered processed text as input. We work on six different machine learning classifiers which produced great accuracy over our dataset. Among six algorithms, Multinomial Naive Bayes and Logistic Regression provided 95% accuracy.