The objective is to acquire Concurrent User Licenses for a Natural Language Processing (NLP) software platform that will to allow data mining of FDA internal documents and external literature sources such as FDA Drug Labels, MEDLINE and other databases. The NLP toolkit is intended to use semantic text mining to rapidly analyze document to find therapeutically useful and possibly unknown associations between drugs, disease processes, adverse events and therapeutic targets.
Over the past five years we have been using the NPL Linguamatics IE2 platform to achieve the goals outlined above. A large number of FDA internal documents and current and archived drug labels have been searched and indexed using Linguamatics IE2. This platform has provided valuable, accurate and reliable information.
The Contractor shall provide access to non-FDA data for search and text mining on their secure cloud environment. The Contractor shall provide federated capability to link multiple servers so that content can reside on different servers in different locations. This is to accommodate the requirement if needed to seamlessly access FDA data on internal servers and standard data sources as listed above from a single user interface. The platform shall have ready built Web Service API’s that allow the FDA to automate indexing of information through scripting or work flow tools. The use of the NLP tool kit shall provide an agile interactive platform that allows users to ask questions against the FDA designated data.
