Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Bill Slawski SEO and the New Search Results

•Download as PPTX, PDF•
•
Bill Slawski
Bill Slawski

Google's search results now include entities and concepts. Entities refer to people, places, things, and 20-30% of queries are for name entities. Google uses meta data like Freebase to build a taxonomy of entities and their relationships. This supports features like the Knowledge Graph, which provides information panels, and allows querying of nearby entities which may soon be available in search results.

1 of 26
SEO and the New Search Results – A look at how Entities and Concepts have grown to become part of Google’s Search Results DFWSEM March 9, 2016 NOAH’s Event Center 2251 N Greenville Ave Richardson, TX 75083
About Bill Slawski • Director of Search Marketing at Go Fish Digital • Founder of SEO by the Sea • Co-Host of Bill and Ammon’s Bogus Hangout
Searchers are looking for Entities… According to an internal study of Microsoft, at least 20-30% of queries submitted to Bing search are simply name entities, and it is reported 71% of queries contain name entities. Building Taxonomy of Web Search Intents for Name Entities
Andrew Hogue Google Senior Staff Engineer and Engineering Manager June 2004 – November 2011: Annotations Framework Structured Data and NLP Strategy Question Answering and NLP Google Squared Google Goggles
Google’s Annotation Framework Andrew Hogue’s Resume
What is an Annotation? Larry Page on the Origins of Google’s PageRank From an entry in Brad DeLong’s journal from 2003. Larry Page: “It wasn’t that we intended to build a search engine. We built a ranking system to deal with annotations. We wanted to annotate the web–build a system so that after you’d viewed a page you could click and see what smart comments other people had about it. But how do you decide who gets to annotate Yahoo? We needed to figure out how to choose which annotations people should look at, which meant that we needed to figure out which other sites contained comments we should classify as authoritative. Hence PageRank. “Only later did we realize that PageRank was much more useful for search than for annotation…” (via @kevinmarks)
Browseable Fact Repository Browseable fact repository https://www.google.com/patents/US7774328 Granted Aug 10, 2010 Filing date Feb 17, 2006 Inventors Andrew W. Hogue, Jonathan T. Betz Original Assignee Google Inc.
Google Acquired Metaweb in 2010
Google’s New Head of Search •Co-Founder of MetaWeb •Former Head of Machine Learning at Google •Inventor of Query Optimization Patent
Query Optimization Patent For example, 9202a8c04000641f8000000000006567, is the GUID which corresponds to the person known as "Arnold Schwarzenegger". The front part, 9202a8c04000641f8, is the database id and the back part, 6567, is the primitive id. Query Optimization US 20110093500 A1, by Scott Meyer, Jutta Degener, Barak Michener, John Giannandrea
MetaWeb Query Optimization
Machine ID Number in Freebase “A mid or machine id is a short form of id for any Freebase topic. Mids allow topics to be more easily managed and changes tracked over time, particularly when they are stored in external data sources and used as foreign keys into Freebase. They differ from a guid in being shorter, and in magically working in spite of topic merges and other such transformations.” ~ http://wiki.freebase.com/wiki/Machine_ID
Type: Body Builder
Type: Actor
Type: Actor
Type: Actor
Type: Politician
Other Uses of MIDs at Google • Reverse Image Search • Trends • Knowledge Graph API • Searches of Nearby Entities (Shortly!)
Reverse Image Search uses MIDs As in ImageNet, the classes were not text strings, but are entities, in our case we use Freebase entities which form the basis of the Knowledge Graph used in Google search. An entity is a way to uniquely identify something in a language-independent way. In English when we encounter the word “jaguar”, it is hard to determine if it represents the animal or the car manufacturer. Entities assign a unique ID to each, removing that ambiguity, in this case “/m/0449p” for the former and “/m/012x34” for the latter. Improving Photo Search: A Step Across the Semantic Gap,
Google Trends uses MIDs to Track* https://www.google.com/trends/explore#q=% 2Fm%2F045c7b (unencoded: “/m/045c7b”) Google Freebase listing (MID /m/045c7b): *(See Barbara Starr slide 26)
MIDs in Google KG Search API A Layman’s Visual Guide To Google’s Knowledge Graph Search API What is JSON-LD? A Talk with Gregg Kellogg
Nearby Entities and MIDs Example data can include a name of the entity, a location of the entity, one or more types assigned to the entity, one or more ratings associated with the entity, a set of entity query patterns associated with the entity, and any other appropriate information that can be provided for the entity. In some examples, an entity can be associated with a unique identifier within the one or more databases. For example, the entity Alcatraz Island can be assigned the identifier /m/0h594. Interpreting User Queries Based on Nearby Locations
The Magic Carpet Ride A statute, near my home in Cardiff-by-the-Sea, known locally as “The Cardiff Kook.” Many Locals don’t know it by its actual name
KG Search API Google’s Knowledge Graph knows both names….
Knowledge Panel Google Shows a Knowledge Panel on a search for its name
We May Be Able to Query Nearby Entities Soon

More Related Content

Bill Slawski SEO and the New Search Results

  • 1. SEO and the New Search Results – A look at how Entities and Concepts have grown to become part of Google’s Search Results DFWSEM March 9, 2016 NOAH’s Event Center 2251 N Greenville Ave Richardson, TX 75083
  • 2. About Bill Slawski • Director of Search Marketing at Go Fish Digital • Founder of SEO by the Sea • Co-Host of Bill and Ammon’s Bogus Hangout
  • 3. Searchers are looking for Entities… According to an internal study of Microsoft, at least 20-30% of queries submitted to Bing search are simply name entities, and it is reported 71% of queries contain name entities. Building Taxonomy of Web Search Intents for Name Entities
  • 4. Andrew Hogue Google Senior Staff Engineer and Engineering Manager June 2004 – November 2011: Annotations Framework Structured Data and NLP Strategy Question Answering and NLP Google Squared Google Goggles
  • 6. What is an Annotation? Larry Page on the Origins of Google’s PageRank From an entry in Brad DeLong’s journal from 2003. Larry Page: “It wasn’t that we intended to build a search engine. We built a ranking system to deal with annotations. We wanted to annotate the web–build a system so that after you’d viewed a page you could click and see what smart comments other people had about it. But how do you decide who gets to annotate Yahoo? We needed to figure out how to choose which annotations people should look at, which meant that we needed to figure out which other sites contained comments we should classify as authoritative. Hence PageRank. “Only later did we realize that PageRank was much more useful for search than for annotation…” (via @kevinmarks)
  • 7. Browseable Fact Repository Browseable fact repository https://www.google.com/patents/US7774328 Granted Aug 10, 2010 Filing date Feb 17, 2006 Inventors Andrew W. Hogue, Jonathan T. Betz Original Assignee Google Inc.
  • 9. Google’s New Head of Search •Co-Founder of MetaWeb •Former Head of Machine Learning at Google •Inventor of Query Optimization Patent
  • 10. Query Optimization Patent For example, 9202a8c04000641f8000000000006567, is the GUID which corresponds to the person known as "Arnold Schwarzenegger". The front part, 9202a8c04000641f8, is the database id and the back part, 6567, is the primitive id. Query Optimization US 20110093500 A1, by Scott Meyer, Jutta Degener, Barak Michener, John Giannandrea
  • 12. Machine ID Number in Freebase “A mid or machine id is a short form of id for any Freebase topic. Mids allow topics to be more easily managed and changes tracked over time, particularly when they are stored in external data sources and used as foreign keys into Freebase. They differ from a guid in being shorter, and in magically working in spite of topic merges and other such transformations.” ~ http://wiki.freebase.com/wiki/Machine_ID
  • 18. Other Uses of MIDs at Google • Reverse Image Search • Trends • Knowledge Graph API • Searches of Nearby Entities (Shortly!)
  • 19. Reverse Image Search uses MIDs As in ImageNet, the classes were not text strings, but are entities, in our case we use Freebase entities which form the basis of the Knowledge Graph used in Google search. An entity is a way to uniquely identify something in a language-independent way. In English when we encounter the word “jaguar”, it is hard to determine if it represents the animal or the car manufacturer. Entities assign a unique ID to each, removing that ambiguity, in this case “/m/0449p” for the former and “/m/012x34” for the latter. Improving Photo Search: A Step Across the Semantic Gap,
  • 20. Google Trends uses MIDs to Track* https://www.google.com/trends/explore#q=% 2Fm%2F045c7b (unencoded: “/m/045c7b”) Google Freebase listing (MID /m/045c7b): *(See Barbara Starr slide 26)
  • 21. MIDs in Google KG Search API A Layman’s Visual Guide To Google’s Knowledge Graph Search API What is JSON-LD? A Talk with Gregg Kellogg
  • 22. Nearby Entities and MIDs Example data can include a name of the entity, a location of the entity, one or more types assigned to the entity, one or more ratings associated with the entity, a set of entity query patterns associated with the entity, and any other appropriate information that can be provided for the entity. In some examples, an entity can be associated with a unique identifier within the one or more databases. For example, the entity Alcatraz Island can be assigned the identifier /m/0h594. Interpreting User Queries Based on Nearby Locations
  • 23. The Magic Carpet Ride A statute, near my home in Cardiff-by-the-Sea, known locally as “The Cardiff Kook.” Many Locals don’t know it by its actual name
  • 24. KG Search API Google’s Knowledge Graph knows both names….
  • 25. Knowledge Panel Google Shows a Knowledge Panel on a search for its name
  • 26. We May Be Able to Query Nearby Entities Soon