Depression is common nowadays without any age limitations. The primary reason for depression is stress. Early detection of stress and treatment will help to cure the same. There are many approaches to detect stress and this paper focus on sentiment analysis. Detection is done based on details in social media and twitter dataset is used. Sentimental analysis is based on extraction from text. Tensi strength is used and this is basically a system for detection. It detects strength of stress as well as relaxation based on expressions posted in Social media. Both positive and negative strength are captured with scaling to indicate the level of stress. Constraint in Tensistrength is it depends on tweets, and drawback here is dis ambiguity. A single word can have different meanings, for example, word Crane, it implies to bird as well as Machine. In order to have better pre-processing SVM with Ngram is used for better accuracy and precision. Our Proposed method works in two phases. During first phase SVM with Ngram is used to detect stress and in the second phase it is fused with Salp Swarm Algorithm. This fusion provides better result compared with PSO.

Stress Detection, SSA algorithm. SVM with Ngram, Sentimental analysis.