farid-1604045/Thesis

Developing a System for Predicting Depression using Machine Learning Technique. Depression is a significant contributor to the overall global burden of diseases. Traditionally, doctors diagnose depressed people face to face via referring to clinical depression criteria. However, more than 70% of the patients would not consult doctors in the early stages of depression, which leads to further deterioration of their conditions. Meanwhile, people are increasingly relying on social media to disclose emotions and share their daily lives, thus social media have successfully been leveraged to help detect physical and mental diseases. Inspired by these, our work aims to make timely depression detection via harvesting social media data. We construct well-labeled depression and non-depression data-set on Facebook and other social media and extract depression-related feature groups covering the clinical depression criteria and online behaviors on social media. A machine learning-based system has been developed for predicting the Depression and preliminary suggestion has been given accordingly. The model has been compared with several machine learning algorithms like Logistic Regression, Decision tree, Naive Bayes, SVM, KNN, Adaboost, Random Forest, CNN, LSTM, BiLSTM, GRU, BiGRU, and Ensemble Learning. The Decision Tree model works best for the Depression data and gives us an accuracy of 98%.