¶ Description
The Sentiment Analysis action allows you to analyze the emotional tone of a given text.
¶ Parameters
¶ Parameters tab
Parameters:
- Developer mode: activate/deactivate code tab.
- Input table partitioning (1st pin)
- Number of lines
- Add button: Add custom parameters
- Trash button: Delete selected parameter(s)
- Key: Specifies the unique identifier of each row in the dataset.
- Column with text to analyze: Points to the column containing the actual text to be analyzed.
- Language column: Specifies the language of the input text using ISO codes (e.g., "en" for English, "de" for German).
¶ Description tab
See dedicated page for more information.
¶ Code tab
sentimentAnalysis is a scripted action. Embedded code is accessible and customizable through this tab.
¶ Configuration tab
See dedicated page for more information.
¶ About
The Sentiment Analysis action allows you to analyze the emotional tone of a given text. It outputs a polarity score (indicating whether the sentiment is positive, neutral, or negative) and a subjectivity score (indicating how subjective or objective the text is). This is particularly useful for text mining, customer feedback analysis, social media monitoring, and other NLP-based workflows.
This action supports multilingual input and dynamically adapts to the specified language column. By identifying sentiment across various contexts, it enables you to gain insight into customer opinions or detect negative/positive trends in large datasets.
NOTE Available languages are:
- "en" (English)
- "fr" (French)
- "it" (Italian)
- "nl" (Deutch)
¶ Input Example
The input table should contain the following columns:
- A unique Key
- A Text column containing the sentences or phrases to analyze
- A Lang column specifying the language of each sentence
¶ Output
The output table will contain the following additional columns:
-
polarity: A score between -1 and 1 where:
-
(-1) indicates very negative sentiment.
-
(0) is neutral, and.
-
(1) indicates very positive sentiment.
-
-
subjectivity: A score between 0 and 1 where:
-
(0) is completely objective.
-
(1) is highly subjective.
Based on the input example, the output might look like this:
-
¶ Example of pipeline :
¶ Use Cases
- Customer Feedback Analysis: Automatically label incoming reviews as positive, negative, or neutral.
- Social Media Monitoring: Track brand sentiment in tweets or posts.
- Product Improvement: Identify areas of concern from negative reviews.
¶ Summary
The Sentiment Analysis action is a powerful tool for extracting emotional and subjective insights from text data. With its multilingual capability and simple setup, it's suitable for real-time analytics and historical data analysis alike. Whether you're exploring customer sentiment or evaluating survey responses, this action provides the foundational layer for emotion-aware data workflows.
