Search Engine Optimization Company and SEO Services - Real SEO Firm

How to solve irrelevance with Textonomy

Any Web user trying to carry out online searching is faced with excessive amounts of irrelevance, incoherence and downright inaccuracy.

Routinely, search-engine enquiries fail to produce relevant results; advertisements fail to link appropriately to websites; and products cannot be found in online store catalogues.

Successful searching can come only from a procedure which places sense at the core of its operation. A sense engine provides the frame of reference within which a search engine can operate. This means relying on those branches of linguistics which analyse what is involved when people handle sense - in particular, semantics and stylistics.

Any procedure which wants to improve the relevance, coherence, and accuracy of online searching has to be semantically based, to enable them to choose the right words to capture relevant records, and also stylistically aware, reflecting the way people vary their language according to their regional, social, and cultural backgrounds.

Semantic and stylistic principles are at the heart of any linguistically based approach to online searching - searchlinguistics.

A lexicological solution derived from this frame of reference is the basis of a suite of products known as Textonomy.

Textonomy rejects the view that searches based solely on statistical algorithms can achieve a successful outcome. It relies instead on the systematic harnessing of human linguistic intuition, tapping into native speakers' knowledge about the semantic relationships between words and the contexts in which words occur.

Because any word in the language can (in principle) be part of a search enquiry, all words have to be analysed to determine their potential as discriminators of online documents.

This is done by assimilating the content of both a dictionary and an encyclopedia The sense engine identifies all the likely search words in a language and rates them for their contextual distinctiveness. The words are then related to a taxonomy of semantic categories that is encyclopedic in scope. Each sense of each dictionary word is assigned an encyclopedia classifications.

The sense engine which drives Textonomy is the result of a searchlinguistics development programme which has taken six years to date and involved an investment of over £4 million in lexicographic and encyclopedic research.

The dictionary component currently has a coverage of over 200,000 items, and the encyclopedia component contains some 4 million words. The apparatus developed to enable the sense engine to function has received a UK patent, with US patent pending.

Source: Crystal Semantics

Posted by nakul at July 20, 2004 10:14 AM | TrackBack