Skip to main content
Production use of this feature is available for specific editions only. Contact our sales team for more information.
The AI Analyze Sentiment transformation component uses the Databricks ai_analyze_sentiment() function to invoke generative AI to perform sentiment analysis on input text. This function uses a Databricks chat model serving endpoints made available by Databricks Foundation Model APIs.
Make sure you have read and understand the Requirements set out by Databricks before using this component.
The returned sentiment score is a string that describes the sentiment with one of the following words:
  • positive
  • negative
  • neutral
  • mixed
If the sentiment can’t be detected, null is returned.

​
Use case

Sentiment analysis classifies the tone or emotional intent of text. This is especially valuable when working with unstructured text fields like customer feedback, reviews, support tickets, or social media content. The output can typically be used with components such as Filter to isolate strongly negative responses, or Aggregate to group sentiment by product, region, or customer. Some typical uses of this component include:
  • Automatically classify sentiment of product reviews, NPS responses, or survey comments, to identify which customers are unhappy and trigger follow-ups.
  • Analyze social media posts to classify brand mentions as positive, negative, or neutral in real-time social feeds and understand brand sentiment.

​
Properties

​
Name
string
required
A human-readable name for the component.
​
Columns
dual listbox
required
Use the arrow buttons or use drag-and-drop to move columns to the right-hand listbox to analyze for sentiment. A new column is created for each column that is analyzed.
​
Include Input Columns
boolean
required
  • Yes: Outputs both your source input columns and the sentiment score columns. This will also include those input columns not selected in Columns.
  • No: Only includes the new sentiment score columns.