Connect Arize AX to Rosette Text Analytics

Both apps 33Arize AX 19Rosette Text Analytics 14

• Trigger
  New Project

  Triggers when a new project is created in Arize AX.

• Action
  Add dataset examples

  Append examples to an existing dataset (CLI: ax datasets append). Endpoint under /v2/datasets/{dataset_id}/examples; confirm in OpenAPI spec.

• Action
  Add prompt

  Create a prompt.

• Action
  Add prompt version

  Create a new version of a prompt (ax prompts create-version).

• Action
  Create Dataset

  Creates a new dataset in Arize AX.

• Action
  Create Experiment

  Creates a new experiment in Arize AX.

• Action
  Create Project

  Creates a new project in Arize AX.

• Action
  Delete dataset

  Delete a dataset.

• Action
  Delete experiment

  Delete an experiment.

• Action
  Delete project

  Delete a project.

• Action
  Find Dataset

  Searches for datasets in Arize AX by ID or name.

• Action
  Find Experiment

  Searches for experiments in Arize AX by ID, name, or project.

• Action
  Find Experiment Runs

  Lists runs for a given experiment in Arize AX. Runs are sorted by insertion order.

• Action
  Find Project

  Searches for projects in Arize AX by ID or name.

• Action
  Get prompt

  Retrieve a prompt by id.

• Action
  List annotation configs

  List annotation configs.

• Action
  List Dataset Examples

  Lists the examples in a dataset, each with its id and its own named values.

• Action
  Update dataset

  Update a dataset's name/description.

• Action
  Update project

  Update a project.

• Action
  Analyze morphology

  Returns morphological analysis of text, such as lemmas, part-of-speech tags, compound components and Han readings.

• Action
  Analyze sentiment

  Analyzes the positive, negative or neutral sentiment of a text, returning the overall document sentiment and the sentiment toward each entity mentioned.

• Action
  Categorize text

  Identifies the contextual categories (QAG Tier 1 taxonomy) of a piece of text or a web page, with a confidence for each.

• Action
  Deduplicate names

  Groups a list of names into clusters of likely duplicates, returning one cluster ID per name in the order supplied.

• Action
  Extract entities

  Extracts the named entities (people, organizations, places and more) from text or a web page, optionally linking them to a knowledge base.

• Action
  Extract events

  Extracts events (such as travel or meetings) from a piece of text, returning each event's type, participants and roles, and whether it was negated.

• Action
  Extract relationships

  Finds the relationships between people, organizations and other entities in a piece of text, returning each subject, predicate, object and a confidence score.

• Action
  Extract topics

  Returns the key phrases and concepts found in a piece of text or a web page, each with a salience score.

• Action
  Identify language

  Identifies the language(s) of a piece of text or web page, returning candidate languages ranked by confidence.

• Action
  Match addresses

  Compares two addresses and returns a similarity score indicating how likely they refer to the same place.

• Action
  Match names

  Compares two names, in any supported language or script, and returns a similarity score showing how likely they refer to the same person, place or organization.

• Action
  Tokenize text

  Splits text into a list of tokens (words and punctuation).

• Action
  Translate name

  Translates a person, place or organization name into another language and script.

• Action
  Transliterate text

  Transliterates Arabic text between Arabizi (Romanized Arabic) and native Arabic script.

No trigger or action matches that. Ask your agent to add it: it reads the API documentation and builds what you describe.