Whitepaper

An excerpt from our whitepaper:

The Semantic Signatures® produced by TSV provide a rich semantic representation of the multiple concepts and topics contained in a text. Semantic Signatures® can be constructed for a wide range of texts, including individual words, phrases, word-lists (e.g. metadata), short passages (such as text advertisements), web pages, or full text documents (e.g. technical articles).

A Semantic Signatures® represents a text through a weighted vector of typically thousands of semantic dimensions. The weight of each vector entry represents the strength of the text along that particular dimension. One can therefore visualize a document as being uniquely positioned in an n-dimensional Euclidean semantic space.

Semantic dimensions are derived in a one-time training process from an appropriate classification schema for the domain. Semantic dimensions may be ‘labeled’ with category names taken from the schema, though in practice dimensions are likely to be used only internally by automated processes.

To request the full whitepaper or if you have additional questions, please contact us.