NltkDetector#

class scikitplot.cleanprompt.NltkDetector(labels=frozenset({'EVENT', 'FAC', 'GPE', 'LANGUAGE', 'LAW', 'LOC', 'NORP', 'ORG', 'PERSON', 'PRODUCT', 'WORK_OF_ART'}), language='en', kind_prefix='', priority=None, confidence=None)[source]#

Detect named entities with NLTK’s maximum-entropy chunker.

Parameters:

labelsiterable of str, optional: Canonical labels to keep. None keeps every label reported. Defaults to DEFAULT_ENTITY_LABELS.
languagestr, default=’en’: Language code. Anything but "en" raises, because the chunker is trained on English only.
kind_prefixstr, default=’’: Prepended to the canonical label to form the placeholder category.
priorityint, optional: Arbitration weight. Defaults to the engine’s declared priority, which is below the structural patterns so an address recognised by both is labelled EMAIL rather than ORG.
confidencefloat, optional: Reported confidence. Defaults to the engine’s declared value, which is lower than spaCy’s — a documented judgement about relative quality, not a calibrated probability.

Raises:

CapabilityError: On first detection, if the tier or its corpora are unavailable, or if a language other than English was requested.

Parameters:

labels (Iterable[str] | None)
language (str)
kind_prefix (str)
priority (int | None)
confidence (float | None)

Examples

>>> detector = NltkDetector()
>>> detector.name
'nltk'

confidence#: !! processed by numpydoc !!

detect(text, policy)[source]#

Yield one span per accepted named entity.

Parameters:

textstr: The original text.
policyRedactionPolicy: The active policy. Unused; present for protocol conformance.

Yields:

Span: Entities whose canonical label is selected, in ascending start order. Every span’s text equals text[start:end].

Raises:

CapabilityError: If the tier, language or corpora are unavailable.
DetectorError: If NLTK raises while processing the text.

Parameters:

text (str)
policy (RedactionPolicy)

Return type:

Iterator[Span]

kind#: !! processed by numpydoc !!

kind_prefix#: !! processed by numpydoc !!

kinds()[source]#

Return the placeholder categories this detector can produce.

Return type:: tuple[str, …]

labels: frozenset[str] | None#: !! processed by numpydoc !!

language#: !! processed by numpydoc !!

name#: !! processed by numpydoc !!

priority#: !! processed by numpydoc !!

reads_view = False#: Whether detect is a pure function of the text it is given, so that it may also read the detection view (CP-098) and have its spans mapped back onto the original. False by default: a detector bound to one document — offsets computed in advance from the original, such as a field or region detector — would have its spans mapped twice and cut through a file’s structure (CP-102). Opt in only when every offset a detector yields indexes the string passed to detect.