Suicide prevention is a critical public health issue worldwide, requiring a multidimensional understanding of risk factors and effective intervention strategies. Psychological recognition is based on social media sentiment analysis where data consists of people’s views and perspectives from various social media platforms. In the recent past, a number of attempts have been made to develop some classification models that can classify the social media posts in the form of texts for early detection of the suicidal cases. This paper develops a new KNN-based classification model in association with VADER named KNN-VADER, which is capable of classifying the social media posts in the context of suicidal tendency. The proposed model is tested with real-life as well as survey dataset. The experimental results show that the proposed model classifies the datasets with more accuracy than the other existing methods.