Bill Slawski posted about a new Google patent application filed in August of 2018 that was just revealed last week named Website Representation Vector to Generate Search Results and Classify Website. This patent can explain a lot around the August 2018 Medic update theories out there.
Here is the abstract of the patent:
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using website representations to generate, store, or both, search results. One of the methods includes receiving data representing each website in a first plurality of websites associated with a first knowledge domain of a plurality of knowledge domains and having a first classification; receiving data representing each website in a second plurality of websites associated with the first knowledge domain and having a second classification; generating a first composite-representation of the first plurality of websites; generating a second composite-representation of the second plurality of websites; receiving a representation of a third website; determining a first difference measure between the first composite-representation and the representation; determining a second difference measure between the second composite-representation and the representation; and based on the first difference measure and the second difference measure, classifying the third website.
In short, this describes how Google can classify the type of web site and then use that to assign maybe different weights to rankings. Bill explained:
The patent application uses Neural Networks to understand patterns and features behind websites to classify those sites.This website classification system refers to “a composite-representation, e.g., vector, for a website classification within a particular knowledge domain.”
Those knowledge domains can be topics such as health, finance, and others. Sites classified in specific knowledge domains can have an advantage in using that classification to return search results as they respond to receiving a search query.
The specific patent document says the "scores may be specific for a particular knowledge domain. For instance, the website classification system 120 can determine multiple queries for a particular knowledge domain within a set of multiple knowledge domains. Some example knowledge domains include artificial intelligence, education, astronomy, and health." In fact, it mentions health, doctors, and medical/medicine over ten times but there is no mention of finance or money - although, it really doesn't matter.
It is pretty facisinating to read this and look back at the theories around E-A-T, YMYL and of course the Medic Update in August 2018, the same time this patent was filed. Of course, Google has said just because Google has a patent, it doesn't mean they use it.
Here are some tweets from the industry:
The Google Medic Update was the first week of August 2018. The same week that Google filed the patent application I wrote about in a post on using neural networks to classify sites based on things like expertise & quality scores: https://t.co/FiA0XE1S0y Interesting timing.
— Bill Slawski ⚓ (@bill_slawski) February 23, 2020
Believing in a coincidence when talking about Google, is like believing that I am a good dancer 🕺🏽 #no 😂
— Gianluca Fiorelli (@gfiorelli1) February 24, 2020
"The patent application uses Neural Networks to understand patterns and features behind websites to classify those sites w/in a particular knowledge domain."
— Glenn Gabe (@glenngabe) February 21, 2020
"Those knowledge domains can be topics such as health, finance, and others." https://t.co/1rKuz8BRLF
Forum discussion at Twitter.
