We propose to develop a software system to generate systematic reviews of the scientific literature on chemical risks. The technology improves on what is currently available because it automates the full process of literature review. Unlike existing technologies that simply help humans find and filter relevant articles, our system uses state-of-the-art neural networks to find, filter, extract, summarize, and aggregate research. Risk assessments, and in particular systematic literature reviews, are an important tool to inform decision making and policy. Our technology will allow the EPA, and other organizations, to make health and environmentally sound decisions based on accurate and up-to-date chemical risk assessments. This in turn leads to saved time and money, and the ability to do much more with the same resources. In addition, because we will carefully validate the technology, we can assure that any assessments developed with it will be high quality, transparent, consistent, and scientifically defensible. Our plan relies on our extensive experience with natural language processing (NLP) and machine learning (ML) systems. In previous work, we developed and commercialized software to automate systematic reviews of clinical studies. In this phase I project our primary objective is to show that this technology can be repurposed and modified to accurately extract chemical risks data from research literature, and to design an end-to-end system that can be implemented in phase II. The technology we develop will be valuable to both government as well as businesses that rely on accurate science-based chemical risk reviews to make decisions. The approach we have planned out will allow us to provide a Software as a Service offering to inform decision making based on the collection and analysis of chemical risk scientific data. Markets that would have a need for our technology include the casualty insurance risk analytics market (which inform underwriting of environmental insurance andrelated lines) and the Environmental, Social and Governance (ESG) data market (which informs investors and Boards about evaluations of the environmental impact of companies). Both of these fast-growing markets are estimated to have an as-yet-unrealized potential market size over $1B.
