Help Center> Natural Language Processing> SDK Reference> Using the LU SDK> Calling LU APIs> Sentiment Analysis (Domain-specific Edition)
View PDF

Sentiment Analysis (Domain-specific Edition)

  • Example request 
    //You are advised to construct and use the client in a singleton pattern to avoid frequent object creation.
NluClient client = new NluClient (AuthMode.AKSK,authInfo);

GeneralSentimentReq req = new GeneralSentimentReq();
req.setContent("Text for analysis");

//Domain type. SentimentDomainConstant.DOMAIN_TYPE_CAR indicates the automotive domain. The value is 2.
//SentimentDomainConstant.DOMAIN_TYPE_E_COMMERCE indicates the e-commerce domain. The value is 1.
req.setType(SentimentDomainConstant.DOMAIN_TYPE_CAR);

try {
    GeneralSentimentResp resp = client.getDomainSentiment(req);
} catch (NlpException e) {
    //A failure is thrown as an exception. For details about exceptions, see section "Exceptions". The client automatically processes exceptions.
}
  • Table 1 describes GeneralSentimentReq parameters.
    Table 1 GeneralSentimentReq parameters

    Parameter

    Mandatory

    Type

    Description

    content

    Yes

    String

    Text to be analyzed. The text must be encoded using UTF-8. Only Chinese text sentiment analysis is supported. If type is set to 1 (e-commerce), the value contains a maximum of 200 characters. If the value exceeds 200 characters, only the first 200 characters are detected. If type is set to 2 (automotive), the value contains a maximum of 400 characters. If the value exceeds 400 characters, only the first 400 characters are detected.

    type

    No

    Integer

    The options are as follows:

    1: E-commerce comments

    2: Automotive comments
  • Table 2 describes GeneralSentimentResp parameters.
    Table 2 GeneralSentimentResp parameters

    Parameter

    Type

    Description

    result

    Result

    Encapsulated object of the returned result
  • Table 3 describes the data structure description of the Result field.
    Table 3 Data structure description of the Result field

    Parameter

    Type

    Description

    content

    String

    Text to be analyzed.

    label

    Integer

    Positive and negative labels are included.

    • 1: positive
    • 0: negative

    confidence

    Float

    Confidence level of a label
Parent topic: Calling LU APIs