Abstract
Interactions among humans on social media often convey intentions behindtheir actions, yielding a psychological language resource for Mental HealthAnalysis (MHA) of online users. The success of Computational IntelligenceTechniques (CIT) for inferring mental illness from such social media resourcespoints to NLP as a lens for causal analysis and perception mining. However, weargue that more consequential and explainable research is required for optimalimpact on clinical psychology practice and personalized mental healthcare. Tobridge this gap, we posit two significant dimensions: (1) Causal analysis toillustrate a cause and effect relationship in the user generated text; (2)Perception mining to infer psychological perspectives of social effects ononline users intentions. Within the scope of Natural Language Processing (NLP),we further explore critical areas of inquiry associated with these twodimensions, specifically through recent advancements in discourse analysis.This position paper guides the community to explore solutions in this space andadvance the state of practice in developing conversational agents for inferringmental health from social media. We advocate for a more explainable approachtoward modeling computational psychology problems through the lens of languageas we observe an increased number of research contributions in dataset andproblem formulation for causal relation extraction and perception enhancementswhile inferring mental states.