I use Google a lot and I am happy with the search results. I am a semiotician, too.
So when I heard that Google is joining the race for the semantic search engine, I couldn't wait, yet I was extremely skeptical. Semantic search is the capability for a computer to understand what you're searching for based on the meaning behind your words (instead of just delivering keyword-based results.
The competitors for the semantic search include Kumo, Microsoft's upcoming redesign of Live Search, Wolfram Alpha (launching in May 09). So far, I have not seen miracles.
I did try the new Google and I found that Google is still the same old Google and an excellent tool for finding information. I have to say that I was not impressed at all. Google is not really delivering a different experience. Semantic search is not only the future of the search, it is part of search at all times, come on, how can you search for information without taking into account its meaning.
As far as my search strategies are concerned, I have a variety of tactics to turn almost ANY search engine into a semantic search champion. A little intuition & imagination - the typical trick such as typing in the beginning of the answer instead of asking a question. Imagine what would write the expert that you are looking for, if you were him. Well, to use this trick, you need to have at least some information already, kindly provided by any key-word search engine. Simple!
2 commentaires:
http://www.cazoodle.com
Cazoodle is providing an excellent form of Semantic Search through its three engines, Apartment, Events, and even Shopping. We are able to give users the chance to find anything in those given categories. So to answer your question, yes, some of us are there.
Semantic search cannot start with only searching in text files with different methods.
The first point to solve is how to put the meaning to the words of text. The best way to do this is to change the text itself and wrap it in some syntax which gives it a meaning.
The second point is that everybody in the world must use the same syntax (meta text) for describing and searching for meaning. There are long trials to build such an ontology, but don't think it is possible at all.
Quicker solution can be use of so called folksomony, which means that the meaning will be put to the text by users.
There is still long way to go for better search, but it is exiting way and every new idea and experience bring us a little step forwards.
Please look also at Freebase (http://www.freebase.com/).
Post a Comment