KTN Indexer

If you need quick and efficient multi-label text classification with a huge number of potentially hierarchically related labels, KTN Indexer is a tool for you.


The KTN Indexer provides automatic text classification and management. It’s primarily intended for media clippings of newspaper and magazine scans, but can be used on any textual data. The system uses efficient machine learning methods to train automatic classifiers, as well as language-specific resources to maximize the classification quality. Moreover KTN Indexer allows for semi-automatic classification assisted by human experts by means of active learning.

Specification & Features
  • Automated text classification capabilities
  • Supports multiple categories (labels) per document
  • Supports hierarchy of categories
  • State-of-the-art machine learning algorithms for classification
  • Language-specific resources to maximize classification accuracy
  • Enables semi-automated classification assisted by human experts/li>
  • Minimizes human annotation effort via active learning

KTN Indexer is a commercial tool. If you are interested in using KTN Indexer, please drop us an email at info@takelab.hr.