Rosette Text AnalyticsRosette Text Analytics

Deepen your data insight with Rosette Text Analytics for RapidMiner Studio by Basis Technology. Rosette enables users to quickly and comprehensively process documents, social media, emails, name lists, and other unstructured data in over 55 Asian, European, and Middle Eastern languages. Better understand your content and customers without leaving the RapidMiner platform.

Leveraging Basis Technology’s twenty-plus years of industry experience, Rosette offers:

Entity, Sentiment, and Topic Analysis

Entity Extraction… locate people, places, organizations, and 15 other entity types

Entity Linking… connect the entities in your text across documents and into the real world

Categorization… classify documents based on the IAB QAG taxonomy

Sentiment Analysis… measure response feedback — positive, negative, or neutral

Base Linguistics

Morphological Analysis… perform linguistic tagging, lemmatization, decompounding, and Han readings

Tokenization… segment words into their component parts

Sentence Tagging… disambiguate sentences in noisy text

Names Analytics

Name Translation… comprehensive, accurate multilingual name translation

Name Matching…  fuzzy, cross-lingual name matching and resolution

Rosette uses natural language processing (NLP)  statistical analysis, and machine learning to power advanced text analysis and names processing.


Signing Up

In order to use the Rosette RapidMiner extension, you will need an API key. The free tier includes up to 10,000 calls per month (up to 1,000 calls/day) and does not require a credit card to sign up.

Get your free Rosette API key today

Once you have your key, you can enter it as a Rosette Connection in RapidMiner Studio.

Check out the quick-start guide on our blog and our interactive documentation to get started. 


Additional Information

Learn more about Rosette for RapidMiner on our website. If you need help, contact support here.

The Rosette Text Toolkit is also available as an on-premise solution. Get in touch with us for more information.



Version 1.8.1

  • Bug fix: UI fix to supported languages drop-down menu for Entity Sentiment


Version 1.8.0

  • New Transliterate operator transforms Arabic text between Arabic chat (Arabizi) and Arabic script.
  • New Deduplicate Names operator returns a "cluster ID" value for each name in a given list of names. You can then sort on "cluster ID" to group together potentially duplicate names. 
  • Extract Entities operator adds an entity linking feature to link extracted entities to their QID in Wikidata.
  • Support for Japanese and French has been added to detect sentiment (negative, neutral, positive) for a whole document or for an entity in a document.
  • Bug fix: Correctly reads value of concurrency count returned by the Rosette API.


Product Details

Version 1.8.1
File size 6.5 MB
Downloads 11578 (8 Today)11578 downloads
Vendor Basis Technology
Category Domain specific operators
Released 10/30/17
Last Update 10/30/17 8:10 PM
License Vendor License
Product web site
Rating 5.0 stars(1)