homedemos & tools › kiffets


a high-precision content classification engine for news companies to curate Long Tail web content



download overview
watch slideshow


The Kiffets Social Indexing Engine combines the perspective of human editors with the scalability of automatic classification. Curators create topically organized information collections called "social indexes."  By analyzing example articles, Kiffets learns curator intent and matches new content from the Web with topics inside an index. By combining human knowledge with AI content models, Kiffets organizes content automatically with much more precision than other indexing technologies.

Kiffets does not require building a complex ontology or close coordination among curators. Instead, each Kiffets curator independently defines indexes and topics according to a natural human-centric point of view.  The system automatically develops relationships across indexes, creating a rich web of connections accessible through its API for publishers.  The result: Kiffets enables readers to discover and explore related topics and articles, increasing their engagement on publishers' sites.

The core technology and REST API for Kiffets are available for licensing to news companies and publishers as software or as a web service.

A demonstration site for personal curation and news aggregation powered by the Kiffets engine is available at www.kiffets.com.