In today’s post I want to get a little bit into the technology behind Google search, especially into the relational algorithm that orders the websites in a search. The idea came to me when i was researching new uses for databases used by great it corporations, like freebase.com. In this particular case freebase is a collection of structured data useful in any domain, the great thing different from other data bases being that users can submit their own researched data bases, transforming it in a wiki portal.
Why I particularly stopped at this website is because the company that founded it, Metaweb was acquired by Google, and so included the whole data base into a knowledge graph, and this is where it becomes interesting. A knowledge graph is a specific to Google knowledge base used to enhance the search results by means of semantic-source information gathered from a wide variety of sources. This implementation allows Google to search for specific question answers in an effort to deliver specific responses, like other answer engines such as Ask Jeeves and Wolfram Alpha. Knowledge bases have been around for time now and are closely related to Big Data processing. The original use of the term knowledge-base described one of the two components of a knowledge-based system. Such a system consists of a knowledge-base that represents facts about the world and an Inference engine that can reason about those facts and use rules and other forms of logic to deduce new facts or highlight inconsistencies. In some ways we can say that because a knowledge based system is different from a more common hierarchical relational base it can be compared with a object oriented model having structures that resemble classes, subclasses and instances.
On a optimistic note, having studied this phenomenon I can only observe the tendency of Google search to become more intuitive, more responsive and most important more accurate.