Hello, Lingo3G!

This is the user's and developer's manual for Carrot Search Lingo3G Clustering Engine version 2.0.0.

What is Lingo3G?

Lingo3G is a text clustering engine — it organizes collections of up to ~10,000 documents into clearly-labeled thematic groups called clusters. You can use Lingo3G to organize, for example, search result into groups like these:

Search results (snippets) and clusters discovered from them, light theme.
Search results (snippets) and clusters discovered from them, dark theme

Search result titles and snippets (on the left) for query "data mining" and corresponding cluster labels (right).

Lingo3G performs clustering fully automatically, based only on the document text you provide. Lingo3G's unique algorithm produces high-quality semantic clusters within fractions of a second.

Is Lingo3G right for me?

You will likely use Lingo3G in one of the following scenarios:

  • As a programming component integrated with the software you develop. In this case, depending on your preference, you will use some of the available Lingo3G APIs:

  • As an end-user app for clustering documents from various sources, such as Excel, CSV or JSON files, public or internal search engines.

    In this case, you will use Lingo3G Clustering Workbench. You will need some IT experience or help to install the app, to prepare your data in the required format or to configure and connect your internal search engine to Lingo3G.

How do I start?

  1. Try Lingo3G demos to see what kind of results you can expect from your data.

  2. Optional: to learn about the basic concepts involved in Lingo3G clustering, have a look at the Basic concepts section.

  3. If you want to use Lingo3G as a programming component, choose the API and then, depending on your decision, proceed to the REST API or Java API section.

  4. If you'd like to use Lingo3G as an end-user app, see Search Results Clustering app and Clustering Workbench app.

  5. Optional: if you'd like to tune the quality or performance of Lingo3G clustering, have a look at Dictionaries and Parameter tuning.