The extraction process

High North Research Documents extracts all relevant documents through a simple but intelligent text mining algorithm. The extraction process is mostly done automatically, but does also involve a manual control.

Our last update resulted in 151 085 automatically extracted records and 5158 we will have to manually control for high North relevance. This work is now in progress and the accepted records will be included in our September update. A preliminary result shows an acceptance rate of 33 %.