These days most people frequently experience stress and anxiety. Chronic and excessive stress can lead to increase in blood pressure, insomnia, heart attacks or even death. Stress has become a prevailing factor for causing mental illness agencies. The Leacock-Chorodow (LCH) algorithm, an advanced deep learning algorithm along with the WordNet library. If we do not get a control on our stress, it becomes deep-rooted and can seriously interfere with our daily activities. So, it is important to detect stress before it interferes with a person’s well-being. Traditional face-to-face psychological diagnosis and treatment cannot meet the demand of peoples' stress completely due to its lack of timeliness and diversity. Nowadays the influence of Facebook, Twitter, YouTube and other social media giants has spread across modern society. People share their daily activities with friends on social media platforms. So, we create a social website where people can interact with their friends and this social media data can be used to analyze user’s stress state. Our model will be useful in developing stress detection tools for health is used to detect the stressed words from a user’s tweet. We subsume two types of attributes namely tweet-level content attributes where we consider each and every tweet or post made by the user and user-scope statistical attribute where the weekly tweet is taken. We propose a Deep Neural Network (DNN) model to incorporate the two types of user-scope attributes to detect users’ psychological stress. Our social website can be used to detect stress based on the user’s interactions with his friends and how active the user is on the social website.