Audience
Companies and law professionals seeking a deep learning solution to manage employee discrimination cases
About Intraspexion
The cost of commercial tort litigation against the business (where benefits paid = losses + defense attorney fees + administrative costs, e.g., gathering documents, meeting with counsel, sitting for depositions, etc.) was about $160 billion per year. For the 10-year period, the total was almost $1.6 trillion. To train a deep learning model of the federal court’s category of civil rights-employment, i.e., “employment discrimination,” we extracted allegations of fact from 400 complaints filed in federal court (no emails). We put the employment discrimination model in a GPU instance in Microsoft Azure and AWS and tested it with 20,401 emails from the Enron. The model had not previously assessed any emails. You can export each true positive email to an internal investigation or case management platform. And, because there’s a database hooked to the UI, you can save those true positives, then add them to the initial training set and re-train the model.