Abstract: Mental health problems leading to depression have become a critical concern due to the growing engagement of people on social media platforms. Several past approaches have been implemented by analyzing the pattern and behaviour of the posts by users on social networking sites. This research study proposed a system for predicting users who may be depressed, based on the characteristics of users who is already affected. A combination of both the tweet-level and the user-level architecture was used to generate a more robust and reliable system where semantic embeddings trained from advanced neural networks were adopted under the tweet-level. SVM with Word2Vec and TF-IDF has been used and yielded an accuracy of 98.14% and recall of 95.63%.