How You’ll Benefit
Automate monitoring and data ingestion from multiple structured and unstructured data sources.
Discover complex relations between entities loaded in a knowledge graph with high quality.
Extend, curate and validate extracted and semantically structured content.
Expose data stored in the knowledge graph via standard APIs to other customer systems.
How it Works
Data collection & crawling
- Key structured data sources for drug products and medical devices – FDA Medical Devices databases, CT.gov, DrugBank, FDA NDC, Drugs@FDA, etc.
- Public research articles and conference proceedings – PubMed abstracts, PMC full text research articles.
- Filed patents from major national patent offices – USPTO, EUPTO, etc.
- Company websites pages and news from industry portals.
Content classification & semantic sectioning
- Classify diverse content into discrete focused subsets.
- Detect document sections to facilitate contextual text extraction.
- Apply focused text analysis pipelines for specific sections and content.
Semantic disambiguation & knowledge graph
- Employ semantic disambiguation based on ontologies and controlled vocabularies.
- Use controlled vocabularies and ontology management tools that facilitate the extension of the terminologies used for text analysis.
- Benefit from a knowledge graph populated from structured data sources and extended with information extracted from the text.