Remediation technology selection is a significant challenge that must considered the types of contaminants present, migration pathways, hydrogeologic conditions, presence of infrastructure, regulatory requirements, and many other factors.
Environmental experts today are exposed to extremely rapid growth of the information (Big Data) making it impossible for an individual to assimilate knowledge available and quickly integrate it to his expertise. Obtaining and processing all appropriate information to diagnose the problem and choose the most efficient solution with high level of confidence is key to an efficient remediation of contaminated sites.
WatRem is a cognitive application designed to help experts in selecting the best remediation technologies for contaminated sites. The application incorporates the new cognitive processing technology created by IBM/Watson to process massive amounts of technical information (Big Data) from a range of sources such as scientific journals, governmental agencies, research centers, case studies, etc. Most importantly, this tool utilizes machine learning techniques to continuously enrich its knowledge over time by integrating the information, as soon as it is available. With the use of IBM/Watson-based cognitive technology (i.e., machine learning and natural language processing), experts can quickly obtain, remediation technology predictions based on reliable scientific data with associated confidence levels.
Primary Author/Conference Presenter:
Daniel Fortin (WikiNet, Quebec, QC, Canada)