SIFTER™ is the proprietary engine that drives DiscoverText, the FOIA Toolkit and PCAT. Our SIFTER™ engine powers many of the functions such as importation, redaction, clustering and classification. Currently, SIFTER™ is a closed-loop system however we certainly have plans to present an API in the future so third-party applications can take advantage of our data and processes.

The SIFTER™ proprietary engine is built upon time-tested programming techniques that scream of reusability and modularity making it easy for the Texifter staff to add and modify functionality and to extend the new behaviors seamlessly into DiscoverText and other Texifter applications. Use of the SIFTER™ service framework is complementary with professional and enterprise licenses for DiscoverText.

SIFTER™ Functions:

  • Background ingestion of archives
  • Background scheduling of live feeds
  • Duplicate detection and removal
  • Near-duplicate clustering
  • Cloud Explorer tag-cloud building
  • Classification of documents
  • Active-learning feedback
  • User and Licensing Systems
  • Other NLP systems coming soon!
  • Future API hooks
  • A common platform for Texifter applications

SIFTER™ at a glance:

Sifter Architecture