Google using phrase based indexing

Google has obtained the patent for the phrase based indexing. This will explore the various documents present in the internet and will find out the frequency with which the same phrase is appearing on the same documents. This will be marked as related by the search engine.

Once this kind of relation marking is done with, clustering will be used to take a decision about the pages that are to be shown in the search results. This will be based on the order of the phrases included in a query by the search engine. It will also depend on the phrases which are closely connected to those particular phrases which are being queried.

So, the order of the pages and the results to be shown in that particular order can be decided using clustering. The system actually looks for the count of the documents which have related phrases, while this clustering is being made use of.

As a result of this kind of searching, we will be looking for the most popular web pages. In fact, the most popular will be returned first and this will also take into consideration, the size of the clusters which are being referred to.