How to realize a Semantic Search

Semantic Web

There are two ways to go about doing this. The first would be "Semantic Web", which would involve making the actual web pages semantic. The second is "Semantic Search Engine", meaning that the software would be built into the search engine, not the web page itself. There are of course pros and cons to each.

The "Semantic Web" would require the creators of a web page to use code to embed the meaning of their text into the page for search engines to read later. This method has actually been around for quite some time and is presumed to be the better of the two. However, this means that it is the individual's responsibility to not only write pages of complex code, but to learn and become fluent in that code, and then enter it into a precise system. Most people are not willing to go through the trouble for this and would rather continue creating their sites using HTML or templates. Therefore, the "Semantic Web" approach is slightly negated by lack of enthusiasm from web creators.


Semantic Searchengines

"Semantic Search Engine" would of course, mean that the semantics are built into the search engine and this would work just as well. However, this method has never actually been tested because no company has had the time nor the resources to try it out. First, an understanding of language similar to that of a human would have to be coded into the engine. Then it would have to be made to run fast enough to sort through thousands of web pages at the click of a button. But by employing this method, it would not be the individual's responsibility and they could continue creating their websites as usual while still being included in the progression towards semantics. This would also mean that existing websites would not have to be re-coded with semantics in order to be search-able.

Leave a Reply

Your email address will not be published. Required fields are marked *