Automatic semantic tagging
Such technology can semantically tag knowledge content across documents so that documents are linked through their knowledge keys. The ability to componentize and highlight document content around major knowledge themes in an organized fashion reflects what you would do if you were to speed-read and highlight pertinent text in documents according to classical speed-reading and learning techniques. Of course, computers can read and organize documents for personalized learning many thousands of times faster than we humans, so this is a tool that we need to leverage to succeed in our world of infoglut.

Automatic user profiling
Knowledge Signatures (which I'll refer to as kSigs) can be bound together along a common semantic context thread to form document collection profiles called Knowledge Threads (which I'll refer to as kThreads, a nary semantic operator). KThreads, by their very nature, cross-associate with other documents inside and outside the thread collection through the themes they have in common within native and external knowledge repositories.

Automatic transformation
KSigs and kThreads can be automatically transformed into powerful and surprisingly accurate fulltext queries, mapped against supported search engines on the Internet, in internal networks and intranets, and across corporate extranets, to retrieve and semantically organize similar documents in real-time for quick learning, decision making, and decision support. You can not only leverage documents for knowledge discovery, but also cut and paste interesting text snippets from documents to retrieve and organize supporting information for targeted learning.

This creates a worldwide virtual knowledge repository at your fingertips, allowing you to quickly power through massive amounts of documents and extract and organize pertinent information without having to learn arcane query languages. The search result documents can then be tested by software transparently and automatically according to their semantic fit with the discovery context. This is done through pattern matching, keeping those documents that fit within its semantic containment, hereafter called the Knowledge Sweetspot.

Automatic knowledge hunting
Furthermore, these results can be shaped into a results vector by common context to the discovery context, document or text, or to the best-fit results document, forming a semantic triangulation that permits the knowledge search engine to recursively reissue more pertinent queries in ever tighter semantics to create a stronger, more semantically dense and relevant query result. This is done through a process called knowledge hunting. Knowledge hunting uses an advanced computational model based on genetic algorithms that emulate LSA (Latent Semantic Analysis), effecting semantic convergence in real-time without the cost of LSA computation.

New results can be snapped to the results vectors according to the shape and density of results required. To make things easier for the client, a picture of a curve that represents an analog of the shape of results could be provided in a browser-based iconographic drag and drop environment so that the engine could respond under the hood automatically in sympathy to user requests.