INDRA (Integrated Network and Dynamical Reasoning Assembler) is an automated model assembly system interfacing with NLP systems and databases to collect knowledge, and through a process of assembly, produce causal graphs and dynamical models.
Running the processor on the latest available data (see https://hershey.dbi.udel.edu/textmining/export/), there are two errors when processing the medline file and they are both about the idString in the id_dict when getting agents containing a badly formatted identifier, both of the form <identifier>;<decimal>', and in both cases MESH identifiers.
Right now the error occurs first at the stage where the statements are checked in the processor with assert_valid_statements.
Suggestions:
Leave the statement extraction as is, but change to check individual statements with assert_valid_statement and skip the invalid statements.
Add a check for semi-colons in where MESH is handled and split on it if present.
Running the processor on the latest available data (see https://hershey.dbi.udel.edu/textmining/export/), there are two errors when processing the medline file and they are both about the
idString
in theid_dict
when getting agents containing a badly formatted identifier, both of the form<identifier>;<decimal>'
, and in both cases MESH identifiers.Right now the error occurs first at the stage where the statements are checked in the processor with
assert_valid_statements
.Suggestions:
assert_valid_statement
and skip the invalid statements.