Integrating AI into P3 Prototype

Hello all,

I attended the American Association of Law Libraries (AALL) conference in Austin earlier this week, and a hot topic for discussion was developments in artificial intelligence (AI). One particularly interesting session involved a presentation by Brian Kuhn, Co-Found and Leader of IBM Watson Legal, and Ed Walters, CEO of Fastcase (a legal research tool, similar to Westlaw or Lexis, focusing on US law). The presentation provided an overview of the services provided by IBM Watson, and how Fastcase is experimenting with implementing IBM’s technology. It was particularly interesting to hear that IBM is opening the technology up to any company that wants to use it for a reasonable licensing fee (more details: https://www.ibm.com/watson). I want to explore how Tologix could deploy the technology in the context of the P3 prototype.

My thinking is that the technology could be leveraged to make our data capture processes more efficient. For example, rather than have subject-matter experts analyze the content directly, they could train a version of the Natural Language Understanding API (https://www.ibm.com/watson/developercloud/natural-language-understanding.html) to analyze material for our purposes and integrate it into our value added tools. This has the potential to greatly increase our capacity to analyze large document sets and expand the scope of the P3 product.

Another example is exploring how the Document Conversion API (https://www.ibm.com/watson/developercloud/document-conversion.html) may be very useful in converting PDF documents to workable HTML files. Anil and his team have already made progress in this regard, but I am interested in exploring how this compares to the IBM technology.

Let's plan on discussing these issues during the project update next Thursday; however, in meantime, please circulate your thoughts on this opportunity.

Thanks,

Morgan