Our digital scientist technology combines knowledge extraction and graph learning

 “Labeled training data is the new new oil.” 
Christopher Ré, Stanford

Together, with world leading experts from Stanford University, we developed a data labeling system to vastly accelerate
the generation of Natural Language Processing (NLP) classifiers for biomedical knowledge extraction.

Building on this technology, OccamzRazor developed the first end-to-end insight prediction pipeline. It allows us to automatically draw knowledge from millions of peer-reviewed publications, high-quality proprietary datasets, clinical trial findings, and our own lab experiments. The seamless integration of text-based and structured datasets in one knowledge map is the basis for our innovative approach to drug discovery.

Our dynamic knowledge map technology acts as a digital scientist and provides new insights for various diseases. The importance of automatically integrating text-based data to find new disease mechanisms and drug targets is a breakthrough finding that we proudly present at the 2020 NeurIPS conference.