openaire-graph-docs/docs/data-provision/enrichment/classified.md

826 B

sidebar_position
5

Classifiers

TODO

Property Description
Short description Classifiers
Authority ATHENA Research Center, Greece
Licence CC-BY/CC-0
Algorithmic details The algorithm classifies publication's fulltexts using a Bayesian classifier and weighted terms according to an offline training phase. The training has been done using the following taxonomies: arxiv, dcc, acm
Parameters Publication's identifier and fulltext
Limitations N/A
Code repository https://github.com/openaire/iis/tree/master/iis-wf/iis-wf-referenceextraction/src/main/resources/eu/dnetlib/iis/wf/referenceextraction
Environment Python, madIS (https://github.com/madgik/madis), APSW (https://github.com/rogerbinns/apsw)