Synonym and Association Generation (used in MT and Search/Text Mining)
One of Meaningful Machines’ NLP (natural language processing) innovations is a “synonym and association generator.” This technology works in any language and simply requires a corpus (body of text) in that language. Given a query term—which can be a word or phrase—the technology has the ability to generate lists of words and phrases (multi-word units) that are synonymous or near-synonymous to the query term or have other direct associations to the query term such as class members, opposites, or descriptions of the term. This technology, which automates the understanding of single and multi-word concepts as expressed in natural language, adds significant power to MT by allowing the MT system to look at and consider synonyms and other related terms during the translation and meaning extraction process. This technology can also be used to enhance search and text mining applications in any language by either expanding or refining search queries. It requires only a corpus of text in the particular language – and it does not use language rules, tagging, stemming or pre-built semantic knowledge of any kind. With a vast majority of the world’s information, including at least 80 percent of all corporate information, tied up in unstructured data (e.g., word documents, e-mails, etc.), this technology enables an automated, unlimited, and intuitive method of enhancing text search, tracking entities and relationships, and improving data management.