You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Redlist descriptions often contain strings describing mechanisms of habitat loss: coral bleaching, pollution, disease, predation, acidification, algal blooming, harvesting...
Many of these correspond to processes in ENVO or other OBO ontologies. It would be useful to use NER to detect these. We could then link these to ENVO-H classes, linking environments to processes that negatively affect them.
This would be an iterative process at first; we would need to gap-fill many processes before this is effective. However if we have the core processes then we can parse post-compositionally (e.g. the string 'destruction' next to 'reef')
(don't have Evangelos' github id so can't tag him here)
The text was updated successfully, but these errors were encountered:
Redlist descriptions often contain strings describing mechanisms of habitat loss: coral bleaching, pollution, disease, predation, acidification, algal blooming, harvesting...
Many of these correspond to processes in ENVO or other OBO ontologies. It would be useful to use NER to detect these. We could then link these to ENVO-H classes, linking environments to processes that negatively affect them.
This would be an iterative process at first; we would need to gap-fill many processes before this is effective. However if we have the core processes then we can parse post-compositionally (e.g. the string 'destruction' next to 'reef')
(don't have Evangelos' github id so can't tag him here)
The text was updated successfully, but these errors were encountered: