Conversational services for multi-agency situational understanding

Abstract

Recent advances in cognitive computing technology, mobile platforms, and context-aware user interfaces have made it possible to envision multi-agency situational understanding as a 'conversational' process involving human and machine agents. This paper presents an integrated approach to information collection, fusion and sense-making founded on the use of natural language (NL) and controlled natural language (CNL) to enable agile human-machine interaction and knowledge management. Examples are drawn mainly from our work in the security and public safety sectors, but the approaches are broadly applicable to other governmental and public sector domains. Key use cases for the approach are highlighted: rapid acquisition of actionable information, low training overhead for non-technical users, and inbuilt support for the generation of explanations of machine-generated outputs.