Traphoty® Rules is a set of customized linguistic rules that can be deployed in two different ways to help overcome the limitations of non-linguistic search solutions.
In its analytical form, Traphoty® Rules can be integrated into a non-linguistic search engine to improve its reliability. This is the product for any organisation which has already deployed a search solution based on similarity keys, and wishes to enhance the reliability and precision of their automated searches by adding linguistic rules.
In its productive form, Traphoty® Rules can enrich data by generating linguistic variants for the names it contains. For example: Traphoty® Rules would detect that the name “Boris Nikolayevich Yeltsin” is of cyrillic origin and would produce alternative transcription variants, such as “Boris Nikolaïevitch Eltsine”, “Boris Nikolajewitsch Jelzin”, and “Bóris Nicoláievitch Iéltsin”. With these variants added, a non-linguistic search engine can be used to search the data set without significant risk of missing false positives due to the lack of linguistic search methods.
The linguistic rule sets which form the heart of the Traphoty® product range can be fully configured to reflect the most appropriate risk profile, and to incorporate specific languages and scripts relevant to your business. If desired, Linguistic Search Solutions consultants are available to assist with this process.
Clients operating regional matching processes can select the most appropriate language rule sets from a range covering all major languages of the world, or request custom rule sets for minor languages, which can be produced on demand by our dedicated team of computational linguists.
