Most search engines offer full text search, i.e. they can
find documents that match the exact wording of the search query. This
produces very limited search results; for example, if you search for
"bicycle" only documents that contain the word "bicycle"
are found, while documents that contain the word "bike," "pushbike,"
etc. are not retrieved. Synonym search, however, finds all of these
related documents. Because it is cross-lingual, InfoCodex's synonym
search goes even one step further. It finds not only all the documents
with the related English words, but also tracks down documents that
contain the words "Fahrrad," "bicicletta," "vélo,"
etc.
Instead of searching for documents which contain a number of keywords,
similarity search finds documents that are similar in contents to a
given query text. To find particular documents, none of the words in
the query text have to be present, only the content of the document
has to be similar. As a result of InfoCodex's cross-language support,
documents in all supported languages (E, F, D, I, ES) are found as long
as their content is similar to the query text.
In both cases, because of the cross-language content recognition and
a well-founded similarity measure, the documents are ordered by their
relevance to the query.