A new Google patent, issued on December 28, 2021, focuses on how to interpret queries and solve them using entity information.
I frequently include references to Google patents in my articles because I spend a lot of time learning from them.
Patents are filed to describe new inventions and encourage potential competitors to innovate. They provide enough information to prevent others in the same industry from copying the patent filers’ intellectual property.
We frequently find interesting information about the assumptions that patent creators make about search, searchers, and the Web, which can make the patents an interesting read as well.
As is customary when I share the highlights of a patent in an article like this, you are encouraged to read the entire patent.
I try to explain what the patent may cover, but not in such a way that my post appears to be a copy of the patent to indexing programs.
You may recall that Google search engineer Paul Haahr gave a presentation on “How Google Works” at SMX 2016.
One of the key takeaways was that Google tries to identify when entities appear in search queries submitted by users.
That statement begs the question of how Google can tell which entity is being referred to in a query.
Google has filed a patent on the subject, which is the subject of this post.
How to Improve Queries Interpretation
Search has evolved to accept such search queries and return relevant results.
Some search engines, however, provide search results without understanding the search query.
In response to [action movie starring Tom Cruise], for example, irrelevant search results such as [Last Action Hero] and [Tom and Jerry] may be returned because a portion of the search query is included in the titles of the pieces of content.
Understanding the search query can assist the search engine in producing more relevant results.
What might a search engine make of a query?
The following methods are mentioned in the patent:
- Receiving a search domain query.
- Choosing search terms based on the query
- Whether a search term corresponds to a name of an entity.
- Check to see if the entity name is from metadata related to the search domain.
- Observing that a large number of entity names correspond to at least a portion of the total number of search terms.
- Choosing an entity type and an entity score for each of the corresponding entity numbers.
- Finding a number of entity names by removing some matching entity names based on the entity score and contextual information in the search query received.
- Performing a search in the search domain using the remainder of the entity names.
- In this case, each entity named in the remaining portion of the number of entity names is searched for in relation to the associated entity type.
This method of query interpretation can also include:
- In a search domain, receiving a voice query.
- Choosing a large number of voice recognition terms based on the voice query received.
- Choosing, for each of the various voice recognition terms.
- Whether or not a portion of a voice recognition term corresponds to a name of an entity.
- Where the entity name is derived from metadata associated with the search domain, and an entity score is assigned to the entity name.
- Using the entity score, calculate a feasibility score for each of the number of voice recognition terms.
- The number of voice recognition terms is ranked based on the feasibility score.
- Choosing one of the ranked voice recognition terms to execute the voice query in the search domain.
This Query Interpretation patent can be found at:
Methods, systems, and media for query interpretation
Yongsung Kim is the inventor, and Google LLC is the assignee.
11,210,289 US Patent
Date granted: December 28th, 2021
5th of May, 2017
Abstract:
Mechanisms for interpreting queries to get provided.
In some implementations, a method for analyzing queries get provided, comprising:
Receiving a search query in a search domain
Determining search terms based on the search query
Determining, for each of the search terms, whether a search term corresponds to an entity name,
Wherein the entity name gets derived from metadata associated with the search domain.
In response to determining that entity names correspond to a part of the search terms
Determining an entity type and an entity score associated with each of the corresponding entity names
Determining a remaining part of the entity names by removing at least one of the matching entity names based on the entity score and contextual information in the search query
Performing a search in the search domain with the remaining part of entity names,
Each entity named in the remaining part of entity names gets searched corresponding to the associated entity type.
The Patent Interpreter’s Concluding Remarks
When a search engine detects the presence of an entity in an article, it will attempt to determine who the entity is.
One Google patent I previously discussed explained how an entity name like “Michael Jackson” may appear to identify only one person with who most people are familiar. He was, after all, a well-known musician and entertainer.
But there was another Michael Jackson who was nothing like the first; he was a Director of Homeland Security.
Google calculates confidence scores to determine which entity may be referred to when it appears in a query.
This patent describes how Google may determine which entity is being searched for before returning results related to that entity.
Remember that when someone searches for “Lincoln” (another Google patent example), they could be looking for a Lincoln town car, Former President Abraham Lincoln, or the city of Lincoln, Nebraska (also many other states).
If the search engine correctly interprets the query and returns relevant results to the searcher, they can meet the searcher’s informational or situational need.
The description of the patent contains a lot more analysis of how this patent works, but I wanted to highlight why it was needed and necessary.
There is too much risk of potential confusion if the search engine does not attempt to correctly interpret a query.
Learn more from SEO and read Responsive Web Design’s Top 7 SEO Benefits.
