Google’s Knowledge Extraction Algorithm – What You Need to Know

Featured image for "What is Google Knowledge Extraction Algorithm" article

What is Google’s Knowledge Extraction Algorithm?

Google’s Knowledge Extraction Algorithm attempts to structure data collected by GoogleBot so that it can annotate the information before putting the information in Google’s Web Index.

The annotation indicates what the information is – an image, a video, a heading, a data table, a paragraph, and aside, a menu etc etc. Additionally, the algorithm attributes score that indicates its level of confidence in the annotation it has given.

Kalicube has identified three Knowledge Algorithms of Google that are responsible for curating information for its Knowledge Graphs and Knowledge Panels.

  1. The Knowledge Extraction Algorithm
  2. The Knowledge Panel Algorithm
  3. The Knowledge Vault Algorithm

The vast majority of data on the web is unstructured, so much of the time, the Knowledge Extraction Algorithm needs to make a best guess and give the information structure. The annotations for this type of information will have a low confidence score.

Some data is structured using Schema.org markup, HTML tables, lists, headings and other structure techniques. The annotations for this type of information will have a high confidence score.

The annotations are used by the ranking algorithms and other Knowledge Algorithms to access the information. The confidence scores are used by those algorithms to assess the reliability of the annotation (and so the confidence score therefore affects the decision-making of the algorithms).

Read more about the Three Knowledge Algorithms of Google here >>

What Are The Three Knowledge Algorithms of Google?

The three Knowledge Algorithms of Google are the Knowledge Extraction Algorithm, Knowledge Panel Algorithm and Knowledge Vault Algorithm.

The Knowledge Extraction Algorithm

The Knowledge Extraction Algorithm annotates the information Googlebot inserts into the web index. It loves structured data: Schema markup, HTML tables, headings and Semantic HTML5 for example. When structure isn’t in place, it attempts to create structured data from (mostly) unstructured online content.

The Knowledge Panel Algorithm

The Knowledge Panel Algorithm builds Knowledge Panels on Google SERPs. The Knowledge Panel Algorithm determines what information it can show that is a factual summary of the Entity. Its ultimate aim is to provide Google’s user with a simple, factual summary about the entity that saves the user the time of clicking on multiple links to research.

The Knowledge Vault Algorithm

The Knowledge Vault Algorithm adds information (hopefully facts) to the Main Knowledge Graph (aka the Knowledge Vault). The ultimate aim for Google is to “understand the world” by filling this  Knowledge Vault with all the facts about everything. Our job is to feed this algorithm with corroborated facts about entities.

Similar Posts