AnalyzeSentimentResult 클래스

AnalyzeSentimentResult는 문서에 대한 전체 예측 감정 및 신뢰도 점수와 점수가 있는 문장별 감정 예측을 포함하는 결과 개체입니다.

상속
azure.ai.textanalytics._dict_mixin.DictMixin
AnalyzeSentimentResult

생성자

AnalyzeSentimentResult(**kwargs: Any)

메서드

get
has_key
items
keys
update
values

get

get(key: str, default: Any | None = None) -> Any

매개 변수

Name Description
key
필수
default
기본값: None

has_key

has_key(k: str) -> bool

매개 변수

Name Description
k
필수

items

items() -> Iterable[Tuple[str, Any]]

keys

keys() -> Iterable[str]

update

update(*args: Any, **kwargs: Any) -> None

values

values() -> Iterable[Any]

특성

confidence_scores

각 감정 레이블에 대해 문서 수준 감정 신뢰도 점수가 0에서 1 사이입니다.

confidence_scores: SentimentConfidenceScores

id

요청과 함께 전달된 문서 ID와 일치하는 비어 있지 않은 고유한 문서 식별자입니다. 요청에 지정하지 않으면 문서에 대한 ID가 할당됩니다.

id: str

is_error

결과 목록을 반복할 때 오류 항목에 대한 부울 검사. AnalyzeSentimentResult의 instance 항상 False입니다.

is_error: Literal[False] = False

kind

텍스트 분석 종류 - "SentimentAnalysis".

kind: Literal['SentimentAnalysis'] = 'SentimentAnalysis'

sentences

문장 수준 감정 분석.

sentences: List[SentenceSentiment]

sentiment

문서에 대한 예측 감정(음수, 중립, 긍정 또는 혼합)입니다. 가능한 값으로는 'positive', 'neutral', 'negative', 'mixed' 등이 있습니다.

sentiment: str

statistics

요청에 show_stats=True 를 지정한 경우 이 필드에 문서 페이로드에 대한 정보가 포함됩니다.

statistics: TextDocumentStatistics | None = None

warnings

문서를 처리하는 동안 경고가 발생했습니다. 경고가 있는 경우에도 결과가 반환되지만 완전히 정확하지 않을 수 있습니다.

warnings: List[TextAnalyticsWarning]