Crystal Semantics has developed the ‘Sense Engine’ in order to produce relevant search results by utilising the senses of words, rather than statistical algorithms used by other search technology. Because any word in the English language can be part of a search enquiry, each word is analysed to determine its potential to discriminate which context the search should cover. The ‘Sense Engine’ identifies all the likely search words, advises the user of the different contexts the search should cover, and categorises the results encyclopedically providing users with results relevant to their request.
The ‘Sense Engine’ is the result of a six-year search linguistics development programme undertaken by Professor David Crystal, a world authority on linguistics, encyclopedia editor and published author for Cambridge University Press and Penguin Books. £4 million has been invested in lexicographical and encyclopedic research, giving the ‘Sense Engine’ a classification system of around 2,000 categories derived from an encyclopedia component of over five million words.
Using the ‘Sense Engine’ users will be able to easily refine their searches to retrieve information that they actually want. For example, a typical search for the word ‘depression’ can return results that relate to mental health, climate, economics, geography, etc, and results will not even be grouped in any logical order. Therefore, if the user is only interested in results in relation to climate they still have spend significant time filtering the other results or expanding their search. In contrast the ‘Sense Engine’ automatically categorises the various senses of ‘depression’ and shows them to the user, allowing them to simply refine the search and only see results relating to ‘economic depression’ for instance.
The ‘Sense Engine’ has immediate applications for corporate/consumer search and navigation on intranet and internet sites, such as classification of documents, online contextual advertising and e-commerce.
David Crystal, chairman of Crystal Semantics, said: “The ‘Sense Engine’ is the next logical step with regards to web searching. Every word in the English language has on average 2.5 related senses, so consequently many users are working harder than they need to in order to search for relevant information. By understanding the semantic context of a search we aim to save users significant time and frustration and enable companies to save money and improve productivity.”
Crystal Semantics has developed Textonomy, the first Sense Engine to deliver significantly more accurate and powerful Internet search results. By applying human “senses” and concepts that current algorithms, semantic systems and other statistical techniques cannot match, Textonomy uniquely uses human linguistic intuition, tapping into the semantic relationships between words and the contexts in which they occur. The Textonomy suite of products includes solutions for search and navigation, e-commerce and contextual advertising.
Kaynak : 