Stressors_Dataset

The data largely includes Twitter posts classified as depressive, collected from some popular open-source datasets. While it is always intended to ensure the credibility and quality of the data sources, it is important to note that the verification process mainly pertains to the dataset itself rather than individual posts or statements within the dataset. Apart from this, a fixed amount of text samples was synthetically generated for each stressor using a large language sequence-to-sequence model. These synthetically generated text samples were used to label the original data collected earlier. This This methodology was adopted due to absence of an open-source large-scale dataset that covers a wide array of everyday stressors.