Depression prediction system from Twitter’s tweet by using sentiment analysis / Nur Amalina Kamaruddin

According to the research conducted by the World Health Organization (WHO) in 2015, approximately 300 million of people around the globe are suffering with depression. The research also shows that there is an increase of 18% in the number depression cases diagnosed between 2007 and 2015. Depression is defined as a mental disorder that leads to constant feeling of sadness and also disintegration of interest in an activity that an individual used to enjoy. It also contributes to the inability to carry out daily activities (WHO, 2015). Thus, a Depression Prediction System was developed to predict depression from tweets. The main function of this system is to classify tweet into “depressed” and “not depressed”. The classification model was built using Naïve Bayes algorithm. The number of data used in this project is 15952 with 1 independent variable and 1 dependent variables. These data in term of tweets need to go through data cleaning and data transformation before it can be processed by the classification model. Once the data has been transformed, it is divided into 80% to be used training data and the remaining 20% as testing data.