Context

Clean-up sites continue to present some of the most complicated environmental challenges in the United States. One of the largest of these sites in located along miles of historic industrial sites along the Passaic River, in Northern New Jersey.  To facilitate a legal discovery process, a Company involved with the Passaic Site project needed to take thousands of reports and documents written over more than 100 years, into a single usable knowledge resource. The Company knew that that is was essentially impossible to perform this work using traditional methods – a team of dozens of paralegals would take more than one year to integrate only a subset of the needed information.  The company turned to Arcurve to help with the document integration and contextualization.

Solution

Arcurve was deployed to ingest decades of historical information in a wide-ranging variety of formats, refining and deciphering all the elements of the data. The information included detailed tables of chemistry data from the various sites over the year, newspaper reports, legal depositions, and historical accounts of industrial activities that impacted the sites. Arcurve consolidated five gigabytes of reports, transcripts, articles, depositions, and other content, and created a single knowledge graph that captured and contextualized almost one hundred years of data.

Once created, this single integrated knowledge resource could be quickly leveraged to identify pages that were relevant to any query.  The Advanced Analytics custom tool facilitated this process; queries could be quickly created that would surface and collect all of the resources, across all the years and terabytes of data, that was most relevant to the question at hand.  Initially the work focused on sources that discussed or included data about specific chemical compounds and chemistry related information. Arcurve has already saved the client hundreds of person-hours of manual investigation – brute force approaches that would have taken immense investment of time and resources have been telescoped into a repeatable process that takes only a few days.