The official Google Blog had a post yesterday about the technologies that are behind the Google ranking, following up from a post about how Google ranks results. Amit Singhal is at it again and he talks about the importance of information retrieval (IR). He goes on to say that Google crawls and indexes pages to understand them and associates important concepts to a page when they're not actually obvious on the page itself.
But what about what the users are looking for? Amit says that to understand queries, Google has incorporated a "best-in-class spelling suggestion system, an advanced synonyms system, and a very strong concept analysis system." Google often can read the typos and give you better options when you misspell words. It is a little more difficult, however, to work with synonyms. Also, Google tries to derive the main concepts from a query.
Finally, Google needs to understand users. Google's motto here is that the "best locally relevant results [are] served globally." Google also utilizes personalization and universal search in this area.
Google also has developed something called Cross Language Information Retrieval (CLIR). According to Singhal:
CLIR allows users to first discover information that is not in their language, and then using Google's translation technology, we make this information accessible. I call this advance: give me what I want in any language.
WebmasterWorld members are suggesting that a lot of what Google is saying here is reflected in many of their SERP changes reports. Specifically, there are changes in "relational term weight," probably as a result of the weight on concepts.
Some people think that Google used this to brag about how far they've come. Yet others are upset that Google didn't even really mention the "technologies" as stated in the title (instead, the article is more focused on results). Also, this article doesn't seem to be particularly groundbreaking.
Still, it's an interesting read.
Forum discussion continues at WebmasterWorld and DigitalPoint Forums.
