Skip to main content

How Google Identifies Related Entities From the Search Query Entity

Google has been granted a patent that explains the process how it identifies related entities after selecting the main entity from the search query itself. Here is how related search query entities are identified whenever a query is entered in Google:

Query Type Identification

There is a separate search engine which classifies a search query into one of the 4 groups:


Search Query Entity Identification

After the classification of the search query, the entity present in the query is identified.  

Fetching Resources Having Relationship With the Main Entity

After Google successful classifies the category and entity name, it find out additional resources that bears a strong relationship with the main entity.

Returning Results Along With Other Related Queries

At the last stage, Google returns resources related to the main entity along with other refined search queries that are related to the main entity which users have searched in the past. 


A search for a query like "Shakespeare" returns results related to William Shakespeare, the literary great along with other related queries like "Shakespeare plays", "Shakespeare poems", Shakespeare's life", "Shakespeare quotes" etc. as displayed below:

Google easily identifies the entity as "Shakespeare" who was an English poet and playwright. It quickly gathers 
resources that bears the closest relationship with Shakespeare and also identifies other terms which the users would be willing to search based on the past search history. 

The Knowledge Graph Results

Now, have a look at the Knowledge Graph results which displays other poets (related entities) as per the main query entity (Shakespeare). The Knowledge Graph plays a vital part here helping Google identify and process queries. 

Full Patent Information Can be Viewed Here: 

Also See: