¶ Description
Sentiment analysis of a sentence.
¶ Parameters
¶ Parameters tab
Parameters:
- Developer mode: activate/deactivate code tab.
- idJoin: Add columns to input table
- idContent: Column containing the text to analyze
¶ Description tab
See dedicated page for more information.
¶ Code tab
vaderSentiment is a scripted action. Embedded code is accessible and customizable through this tab.
¶ Configuration tab
See dedicated page for more information.
¶ About
The vaderSentiment action button performs sentiment analysis on text data, determining:
- Positivity
- Neutrality
- Negativity
It returns sentiment scores for each input sentence.
¶ Sample Input Data (CSV)
ID,Text
1,"I absolutely love this product, it's fantastic!"
2,"This is the worst service I've ever experienced."
3,"The weather is okay today, not too bad."
4,"I'm not sure how I feel about this."
5,"The movie was neither good nor bad, just average."
¶ Pipeline Example
The pipeline structure:
¶ Example Output (Data Table)
Notes
- Uses the VADER Sentiment Analysis algorithm.
- Works best for English text.
- Suitable for analyzing short sentences, social media posts, or customer feedback.
¶ Usage Tips
-
Best suited for short text such as:
- Customer reviews
- Social media posts
- Product comments
- Emails or chat messages
-
Interpretation:
sentimentscore: Overall sentiment (-1 to +1 scale)positivity,neutrality,negativity: Individual sentiment components (0 to 1 scale)
-
For neutral or mixed texts, the sentiment score may lean towards zero with higher neutrality.
-
For most accurate results, ensure text is in English.
¶ Troubleshooting
| Issue | Solution |
|---|---|
| Output sentiment scores seem unexpected | Ensure the text data is clean, without excessive symbols. |
| Text with mixed languages produces odd results | Only English is fully supported; non-English text may fail. |
| Action button doesn’t produce output | Confirm that idContent parameter points to a valid text column. |
| Error about missing columns | Check column names and ensure correct mapping in parameters. |
¶ Known Limitations
- The model doesn't detect sarcasm or advanced context shifts.
- Works best for shorter sentences; long text may dilute results.
- It doesn't support deep contextual understanding like modern transformer-based models.
