Nowadays, governments are adopting Web 2.0 technologies for interacting with citizens, empowering them to share their views, react to issues of their concern and form opinion. In particular, social media play an important role in this context, due to their widespread use. For governments, a major technical challenge is the lack of automated intelligent tools for processing citizens’ opinion in government social media. At the same time, during the last decade, argumentation theory has consolidated itself in Artificial Intelligence as a new paradigm for modeling common sense reasoning, with application in several areas, such as legal reasoning, multi-agent systems, and decision support systems, among others. This paper outlines an argument-based approach for overcoming such challenge, combined with context-based information retrieval. Our ultimate aim is to combine context-based search and argumentation in a collaborative framework for managing (retrieving and publishing) service- and policy-related information in government-use social media tools.