RESUMEN
Loneliness is a global public health issue, but the dynamics of loneliness are not understood. Through a global loneliness map, we plan to understand the dynamics of loneliness better by analyzing social media data on loneliness through social intelligence analysis. In this paper, we present the first proof of concept of the global loneliness map. Data on loneliness using keywords associated with loneliness was collected from the USA and analyzed to find meaningful associations of themes with loneliness. The NLP tool used for sentiment analysis of the tweets is a valence aware dictionary for sentiment reasoning (VADER). The tweets with negative sentiment were further analyzed for psychosocial linguistic features to find meaningful correlation between loneliness and socioeconomic and emotional themes and factors. Loneliness is subjective, hence social intelligence analysis through social media and machine learning tools can help us better understand loneliness.