run NER/CR over all textual metadata fields

INCATools / biosample-analysis

analysis of biosamples in INSDC

3 stars 1 forks source link

run NER/CR over all textual metadata fields #31

Open cmungall opened 4 years ago

cmungall commented 4 years ago

Execute RUNner over

study table (see #1)
sample table

Run over all textual fields, in particular:

description fields
envo triad fiels

Vocabularies: ENVO, CHEBI, NCBITaxon in text fields, specifically

this can then be used to repair the tsv to insert the correct identifier for the ENVO class; also for prediction

hrshdhgd commented 3 years ago

I have completed my first pass at running NER against the metadata. Note I have only used ENVO as a dictionary for starters. We could add more ontologies as deemed necessary. The output can be downloaded from here.

@cmungall ; @wdduncan ; @realmarcin