UNSILO is a suite of technologies for knowledge representation and text analytics. UNSILO extracts the most important semantic concepts from a document and it learns and comprehends the topics, things, and events that connect documents in a content collection. The techology is domain-independent; UNSILO can learn the key concepts and ideas in a completely unknown knowledge domain without human guidance or access to taxonomies or ontologies. 

During concept linking, the UNSILO pipeline identifies references to existing ontology and taxonomy terms and resolves any ambiguous references to the most likely matching entity using a mathematical knowledge model. The pipeline is complemented with Dynamic NER models that can detect previously unseen chemical entities, mathematical formulas, software code references, and names of people, places, and organizations.

The UNSILO technology is packaged in products for academic publishing that help publishers to classify content; screen and evaluate incoming manuscript submissions; find the best matching peer reviewers, by comparing the semantic fingerprint of a manuscript with recent published articles in the same subject; match incomings manuscripts with journals; compare manuscripts with already published research to e.g. determine potentially missing references or potential plagiarism.