One of Google's past patents seems to me to describe what they could use, Industry Email List or something quite similar, and it's based on machine learning. It's called "website quality signal generation," and it was noted by Bill Slawski in a brief summary and commentary in 2013 when the patent was granted. It describes how humans could be used to rate the quality of websites, and then an analytics algorithm could match those ratings to website signals — likely Industry Email List automatically identifying relationships between quantified signals and human rating values — and thus generating models from the characteristic signals.
These signal patterns could then be used to compare Industry Email List with other other unranked websites to apply a quality score to them. The wording is quite fascinating: “Reviewers (i.e. people) connect to the internet to Industry Email List view websites and rate the quality of each of the websites. Reviewers can submit website quality ratings to the Quality Analysis Server through the Rating Entry Device. The quality analysis server receives the website quality ratings from the rating input devices and stores the website quality ratings in the signal store.
Website quality ratings are associated with a uniform Industry Email List resource locator and other website signals corresponding to the rated website. The Quality Analytics Server identifies relationships between website quality ratings and website signals and creates a model representing the relationships, as described below. In addition, the quality analysis server searches the signal memory for websites that are not rated (for example, sites with no signal indicating a quality rating). Industry Email List The quality analysis server determines whether unrated websites have website signals related to quality ratings.