Perform Text Analysis and Translation with a Language Action

You can use AI to perform in-depth text analysis and machine translation with a language action. Oracle Cloud Infrastructure Language enables you to process unstructured text for sentiment analysis, entity recognition, classification, translation, and more.

Capabilities

OCI Language is a cloud-based AI service that enables you to build intelligent applications by using REST APIs and SDKs to process unstructured text for language detection, text classification, recognition of named entities, key phrase extraction, sentiment analysis, text translation, and detection of personal identifiable information. OCI Language can identify more than 100 languages in text. It also automatically recognizes at least 18 entity types, including the names of organizations and products. It makes text analysis easier for large volumes of text data. Oracle Integration supports using OCI Language in an integration with the language action.

See AI Language.

Prerequisites

See Prerequisites for information on the prerequisites you must satisfy in the Oracle Cloud Infrastructure Console.

Invoke Oracle Cloud Infrastructure Language from an Integration

  1. Add a Language action to an integration in either of the following ways:
    • On the side of the canvas, click Actions Integration actions icon and drag the OCI Language action to the appropriate location.
    • Click Add icon at the location where you want to add the OCI Language action, then select OCI Language.
  2. Enter a name and optional description.
  3. Select the action you want to perform.
    Element Description
    Action
    • Language detection: Detects languages based on the text provided, and includes a confidence score.

      OCI Language detects the language and returns the detected language along with a related confidence score (between 0 and 1). You can also specify a batch of records.

    • Named entity recognition: Identifies common entities, people, places, locations, email, and so on.

      OCI Language extracts the entities in the text records. It returns the type/subtype and confidence score (between 0 and 1) for each entity.

    • Key phrase extraction: Extracts an important set of phrases from a block of text.

      OCI Language extracts the key phrases from the text. For each key phrase, it returns a score (between 0 and 1) that highlights the importance of the key phrase in the context of the text.
    • Sentiment analysis: Identifies the tone of the text and classifies the expressions in the text into positive, negative, neutral, or mixed polarity.

      OCI Language supports both aspect-based and sentence-based sentiment analysis. For example, opinions, appraisals, emotions, or attitudes toward a topic, person, or entity. After the analysis, it returns a confidence score for each of the classes (positive, negative, neutral, or mixed).

    • Personal identifiable Information (PII): Identifies, classifies, and de-identifies private information in unstructured text.

      OCI Language helps identify and classify personal identifiable information such as name, age, address, email, telephone number, and so on. It returns the information that was identified and classified.

    • Text classification: Identifies the document category and subcategory that the text belongs to.

      OCI Language analyzes the text and automatically classifies it into a set of predetermined categories and sub-categories. It returns this information for each record that was classified.

    • Text translation: Translates text into the language of your choice.

      OCI Language translates the text you provide from the source language to the specified language. It returns the translated text.

  4. Click Continue.
  5. Select the following information, then click Continue.
    Element Description
    Compartment

    Select the Oracle Cloud Infrastructure compartment in which your Oracle Integration is installed.

  6. On the Summary page, click Finish.
  7. Open the mapper and complete the configuration.
  8. Expand the topmost node in the Target section.
  9. Within that node, expand Request Wrapper, expand Body, and then expand Documents.
  10. Right-click Key, and select Create target node.

  11. Click Design View Switch view icon in the Expression Builder.
  12. In the Expression Builder, specify a value for Key.
  13. In the Target section, within Documents, right-click Text, and select Create target node.
  14. In the Expression Builder, specify the text on which you want to perform the OCI Language action you selected in Step 3.

    Note:

    You can optionally specify Compartment Id in the mapper to override the value you selected initially for Compartment in Step 5.


    The Sources, Mapping canvas, and Target sections are shown. The target Key element and the target Text element are set.

    If you selected the Sentiment analysis action in Step 3, then set the Level target element to SENTENCE in the expression builder for sentence level validation.


    The Sources, Mapping canvas, and Target sections are shown. The target Level element is set to SENTENCE.

    If you selected the Text translation action in Step 3, then you need to specify the source language code using the Language Code target element and the target language code using the Target Language Code target element. For example, to translate text from English to French, set Language Code to en and set Target Language Code to fr.


    The Sources, Mapping canvas, and Target sections are shown. The Key element, Text element, Language Code element and Target Language Code element are set in the Target section.

  15. Exit the mapper.

    The language action is now configured.