Recognizes named entities (person and company names, etc.) in text.
The set of entities recognized is language-dependent, and for other
languages, the recognized set is frequently more limited than what is
described below for English. As the name “NERClassifierCombiner”
implies, commonly this annotator will run several named entity
recognizers and then combine their results.
For English, this annotator recognizes
named (PERSON, LOCATION, ORGANIZATION, MISC), numerical (MONEY,
NUMBER, ORDINAL, PERCENT), and temporal (DATE, TIME, DURATION, SET)
entities. Named entities are recognized using a combination of three
CRF sequence taggers trained on various corpora, such as ACE and
MUC. Numerical entities are recognized using a rule-based
system. Numerical entities that require normalization, e.g., dates,
are normalized to NormalizedNamedEntityTagAnnotation.
It is possible to recognize additional or more fine-grained entity
classes through the use of TokensRegex patterns. See also the
||Annotator class name
||NamedEntityTagAnnotation and NormalizedNamedEntityTagAnnotation
||Whether or not to use SUTime. (On by default in the version which includes sutime, off by default in the version that doesn’t. If not processing English, make sure to set this to false.
||A comma-separated list of NER model names (or just a single name is okay). If none are specified, a default list of English models is used (3class, 7class, and MISCclass, in that order). The names will be looked for as classpath resources, filenames, or URLs.
||Whether or not to use numeric classifiers, including SUTime. These are hardcoded for English, so if using a different language, this should be set to false.
||Tells SUTime whether to mark phrases such as “From January to March” as a range, instead of marking “January” and “March” separately.
||If marking time ranges, set the time range in the TIMEX output from SUTime.
StanfordCoreNLP includes SUTime, Stanford’s temporal expression
recognizer. SUTime is transparently called from the “ner” annotator,
so no configuration is necessary. Furthermore, the “cleanxml”
annotator now extracts the reference date for a given XML document, so
relative dates, e.g., “yesterday”, are transparently normalized with
no configuration necessary.
SUTime supports the same annotations as before, i.e.,
NamedEntityTagAnnotation is set with the label of the numeric entity (DATE,
TIME, DURATION, MONEY, PERCENT, or NUMBER) and
NormalizedNamedEntityTagAnnotation is set to the value of the normalized
temporal expression. Note that NormalizedNamedEntityTagAnnotation now
follows the TIMEX3 standard, rather than Stanford’s internal representation,
e.g., “2010-01-01” for the string “January 1, 2010”, rather than “20100101”.
Also, SUTime now sets the TimexAnnotation key to an
edu.stanford.nlp.time.Timex object, which contains the complete list of
TIMEX3 fields for the corresponding expressions, such as “val”, “alt_val”,
“type”, “tid”. This might be useful to developers interested in recovering
complete TIMEX3 expressions.
Reference dates are by default extracted from the “datetime” and
“date” tags in an xml document. To set a different set of tags to
use, use the clean.datetags property. When using the API, reference
dates can be added to an
although note that when processing an xml document, the cleanxml
annotator will overwrite the
“datetime” or “date” are specified in the document.
It is possible to run StanfordCoreNLP with NER
models that ignore capitalization. We have trained models like this
for English. You can find details on the
Caseless models page.
For more details on the CRF tagger see this page.