Loving the Data Others Don’t

The Challenge of Datamining

Like it or loathe it, plain text is a goldmine of information. The challenge is that data mining is often complicated by ambiguity. Sure, identifying, disambiguating and extracting those scientific terms is a big challenge – but we’ve got it covered.

Fit for purpose

Our 80+ hand-curated vocabularies containing more than 20 million synonyms, enriched over any publicly available alternative, are just what’s needed for the job.

Combined with our entity extraction engine analysing over 1 million words per second (that’s the entire Harry Potter collection every second), you have a pretty powerful solution in front of you.

The end result

Taking Medline as an example: over 24 million articles of unstructured plain text. Using just 6 vocabularies we identified 121 million individual, disambiguated assertions in 4.5 hours (did we mention our tools are fast?) and that was just on a laptop.

Back to the topic of data mining, learn more about our text analysis engine.