Possibly more than in any other industry, pharmaceutical R&D is governed by trends and risks. What’s currently hot, what the competition is doing, and what the latest scientific advances are, all decide between a commercially successful medicine and a very expensive dud. There is a tonne of information - perhaps too much - and overlooking a key piece can be detrimental. Furthermore, sifting through mountains of publications is a huge burden, costing both time and resources.
After the pharma patent information is read into a KNIME workflow, a model is trained to identify and tag “interesting compounds”. The workflow is deployed on KNIME Server as a Guided Analytics Application where Competitive Intelligence Managers interact with an Analytical Application. Here they are able to manage and triage the flood of incoming patent data. Tags are drawn from the patent text and “interesting compounds” are identified immediately. The most interesting ones are sorted to the top of the pile and interactive reports are generated.
Why KNIME Software
Data science teams use their expertise to score and categorize the patent information using KNIME Analytics Platform. Competitive Intelligence Managers and Bench Scientists get to access the data quickly and independently via an Analytical Application. Subsequent consumers of the information can also easily update or enhance the entries using the interactive touch points in the KNIME WebPortal.