Google’s Knowledge Vault Algorithm – What You Need to Know
What is Google’s Knowledge Vault Algorithm?
The Knowledge Vault Algorithm adds fact to the Main Knowledge Graph (aka the Knowledge Vault).
The ultimate aim for Google is to “understand the world” by filling its Knowledge Vault with all the facts about everything. Obviously impossible, but a nice target to aim for. Since Google owns multiple human curated Vertical Knowledge Graphs, one technique it is using is to move the entities from the verticals and into the Main Knowledge Graph.
As such, this Knowledge Algorithm is responsible for selecting the entities or information it can move from the different Vertical Knowledge Graphs (Podcasts, Books, Scholar etc), to the Main Knowledge Graph.
This process is pure Machine Learning (ie the algorithms decide, and humans don’t get a say!). It requires that the algorithms assess multiple resources from around the web that corroborate information in the knowledge vertical to a level where it is absolutely sure the information is true.
So, every entity and piece of information that is in the Main Knowledge Graph has been thoroughly checked and confirmed by the Algorithm beforehand. Some sources can skip this validation process. Wikipedia is one example. However, that is less and less the case.
Once an Entity is in the Knowledge Vault (Main Knowledge Graph), it is attributed a confidence score. Over time, if the Knowledge Vault Algorithm receives corroboration from multiple authoritative sources, it will increase the confidence score. Kalicube Pro has recorded confidence scores as low as 0.0001 and has high as 729,927. As of October 2022, the highest Confidence Score is 597,097 (Amazon Web Services), with Google Cloud in second place with a Confidence Score of 486,820.
In the Knowledge Panel Course in the Kalicube Academy we cover how this Knowledge Vault Algorithm functions and strategies you can use to get an Entity into the Knowledge Vault and build the confidence score.