Hello readers,
In the next few posts I will explain “Why there is a growing need for a semantic search?” and “Why semantic search is far better than key word search?”
A good example for the semantic power that can be achieved by search engines is utilized by Setrue Patent Search, this engine is one of the best in the world and its ability to find patents in a very demanding environment of patent searching is remarkable. It was developed by Transformer software Ltd. Setrue Patent search holds two world records in precision score.
The early days
As early as the first days of the invention of computers, people used their computers to store their documents, and in those early days the problem of retrieving documents has began but left unnoticed by information researchers because of the low amounts of information needed to be retrieved. For example, consider a situation when the internet contained only 10 documents altogether (a true situation), if researchers were to develop a primitive algorithm to retrieve information from those 10 documents, this algorithm will always have a precision score of 100% because the probability of finding one document in a set of 10 when given a set of ten result is 10/10 = 100%. Now lets move a bit forward in time to the situation when the internet contained 1000 documents, the same algorithm used before will now yield a score of 10/1000 = 0.1%. That was the point when information researchers realized that they have a real problem.
Stay tuned for the next post…
Tags: patent, patent search, semantic search
