Big Open Source Social Science (BOSSS)

The team of Boston Fusion Corp., Arizona State University, and Sophia Speira proposes Phase II of Big Open-Source Social Science, BOSSS, a technology that will provide rapid social-science based awareness of the cultural landscape in a given locality. Based on the twin pillars of mathematical social science and big data, BOSSS searches open-source documents, parses the documents using natural language processing, or NLP, to fill a knowledge graph, and then analyzes the knowledge graph to draw conclusions based on social science principles. In Phase I, we developed an end-to-end prototype and tested it for two use cases. In Phase II, we propose to build on this foundation by enhancing the robustness and accuracy of the NLP modules, supporting source documents in languages other than English, using social science theory to answer additional questions and to assess the quality and significance of its conclusions, and to set up a professional-quality user interface and API to prepare for transition or integration into a larger framework. We expect that BOSSS will provide actionable social science inputs with neither the long timescale that a human social scientist requires nor the blunders that result from ignoring decades of work in social science theory.