Session: Towards the Utilization of Social Media for Intervention Development: Insights from Domain Expertise, Natural Language Processing and Machine Learning (Society for Social Work and Research 23rd Annual Conference - Ending Gender Based, Family and Community Violence)

Social media are used by individuals, groups, organizations and businesses around the world to create and maintain networked connections. These connections go far beyond family and friends: concerned citizens, nonprofit organizations, and even gangs are engaged in disseminating their story through social media platforms. These weak, but interpersonal ties, result in users voluntarily posting information on their policy positions, emotional states, illegal or violent activities, grief and advocacy campaigns.

With the right methodological tools, social media may support building more efficacious interventions across social work domains.

This symposium brings together four papers using domain expertise, natural language processing and machine learning to showcase the power of interdisciplinary collaborations with computer science and the types of insights that can be gleaned from social media data in social work research. The symposium aims to:

1. Showcase personal/interpersonal insights from social media data, useful for intervention development 2. Introduce the audience to natural language processing and machine learning methods for unstructured text data analysis 3. Underscore the importance of domain expertise in contextualizing natural language processing algorithms.

The generally unsolicited nature of social media data makes it uniquely able to contribute to the social work body of knowledge on lived experience, from multiple positionalities, that could result in more inclusive, efficacious interventions. This symposium offers a set of foundational approaches to building the toolbox towards this goal.