sia reads documents
using linguistic algorithms
A forensic linguist analyses documents by assessing the choice of words, spelling of words, their order, and style of use. When multiple documents are compared using these principles the forensic linguist can find document similarity, common terms and authorship.
sia is based on principles used by
forensic linguists
how sia works - 3 steps
1
automatically create search terms from source document or text
SIA takes the source (search) document and analyses how it is written, breaks it down into sentences, multiple logical word joins, and single words. The system’s algorithms provide a detailed matrix of search terms are then used for comparison to other documents for similarities.
2
rapidly compare search terms to very large document sets
Similarities are determined by similar use of language, sentence makeup, and how words are joined together including removing unwanted non-content words. This information is used to arrive at a similarity match for the documents
3
ranked results that clearly show why they are similar
SIA shows results in ranked order clearly highlighting similarities between the search and reference documents.