At one point in time, search engines such as Google learned about topics on the Web from sources such as Yahoo! and the Open Directory Project, which provided categories of sites within directories that people could skim through to find something that they might be interested in.
Those categories included hierarchical topics and subtopics, but human beings managed them, and both directories have closed down.
In addition to learning about categories and topics from such places, search engines used such sources to do focused crawls of the web to make sure that they were indexing as wide a range of topics as possible.
It’s possible that we are seeing those sites replaced by sources such as Wikipedia and Wikidata and Google’s Knowledge Graph and the Microsoft Concept Graph.
Google may start showing topical search results using breadcrumbs in the search results they return to queries to illustrate topics.
Last year, I wrote a post called, Google Patents Context Vectors to Improve Search. It focused upon a Google patent titled User-context-based search engine.
In that patent, we learned that Google was using information from knowledge bases (sources such as Yahoo Finance, IMDB, Wikipedia, and other data-rich and well-organized places) to learn about words that may have more than one meaning.
An example from that patent was that the word “horse” has different meanings in different contexts.
To an equestrian, a horse is an animal. To a carpenter, a horse is a work tool when they do carpentry. To a gymnast, a horse is a piece of equipment that they perform maneuvers upon during competitions with other gymnasts.
A context vector takes these different meanings from knowledge bases and the number of times they are mentioned in those places to catalog how often they are used in which context.
I thought that knowing about context vectors was useful for doing keyword research. Still, I was excited to see another patent from Google appear where the word “context” played a featured role in the patent. When you search for something such as a “horse,” the search results you receive will be mixed with horses of different types, depending upon the meaning. As this new patent tells us about topical search results:
The ranked list of search results may include search results associated with a topic that the user does not find useful and/or did not intend to be included within the ranked list of search results.
If I was searching for a horse of the animal type, I might include another word in my query that identified the context of my search better. The inventors of this new patent seem to have a similar idea. The patent mentions
In yet another possible implementation, a system may include one or more server devices to receive a search query and context information associated with a document identified by the client; obtain search results based on the search query, the search results identifying documents relevant to the search query; analyze the context information to identify content; and generate a group of first scores for a hierarchy of topics, each first score, of the group of first scores, corresponding to a respective measure of relevance of each topic, of the hierarchy of topics, to the content.
From the pictures that accompany the patent, it looks like this context information is in Headings that appear above each search result that identifies Context information that those results fit within, providing searchers with topical search results. For example, here’s a drawing from the patent showing off topical search results (showing rock/music and geology/rocks):
This Topical Search results patent does remind me of the context vector patent, and the two processes in these two patents look like they could work together. The patent also shows how topics may be organized.
This patent is:
Context-based filtering of search results
Inventors: Sarveshwar Duddu, Kuntal Loya, Minh Tue Vo Thanh, and Thorsten Brants
Assignee: Google Inc.
US Patent: 9,779,139
Granted: October 3, 2017
Filed: March 15, 2016
Abstract
A server is configured to receive, from a client, a query and context information associated with a document; obtain search results, based on the query, that identifies documents relevant to the query; analyze the context information to identify content; generate first scores for a hierarchy of topics, that correspond to measures of the relevance of the topics to the content; select a topic that is most relevant to the context information when the topic is associated with a greatest first score; generate second scores for the search results that correspond to measures of relevance, of the search results, to the topic; select one or more of the search results as being most relevant to the topic when the search results are associated with one or more greatest second scores; generate a search result document that includes the selected search results; and send, to a client, the search result document.
It will be exciting to see topical search results start appearing at Google.
This filter must work in conjunction with context-based Rank Brain and measure the relevance of search terms for assigning high ranking scores.
Google changes so much every single day… I am already very scared of the infinite knowledge box and now there is more to worry. Sometimes I think I should have chosen any other carrier other than SEO.
Anyhow love your blog, Pretty awesome minimal design now it has. I love it.
Hi Cathie,
Since this patent talks about learning about topics from sources such as Yahoo and ODP, I suspect the technology behind it is a little more dated than Rankbrain. The Context Vector patent learns about Contest from Knowledge panels, so it is a little more up to date, and may be built with RankBrain in mind.
Hi Sundar,
There is always something new to learn everywhere, in all kinds of things. I think if you set aside a little time everyday to learn something new, it will be manageable for you.
Sundar is right, Google is changing all the time and that leaves one with no idea of what exactly they are looking for. SEO is really hard and more time consuming.
google is always changing and they are very focused on machine learning. From venice, panda, penguin and last EMD Update. They are making it harder.
Nice share.
Hi Ashish,
I don’t see this patent targeting making anything more difficult for siteowners, but paying more attention to including words that help express context better, so that you page would get ideally labeled would be a good idea.
Hi Tshire, Google does change almost every day. They have told us that they do average around two changes to their core ranking algorithm every day. But, if you follow good information retrieval practices and well know best practices, you should still have a decent idea of what works well to get pages to rank well in search results.
I do look at newly granted and newly published patents every week, and while there are some interesting changes, it is possible to keep up with them.
Hi Bill,
I’ve followed you and the SEO by the Sea blog for a long-time, but rarely comment so “hello!” first of all.
You mention using context vectors to help with keyword research, is there any chance you could expand a little on that – perhaps in a new post? Keyword research is something that I have done for my own SEO clients for the past 5 years, but in truth, my methods have rarely changed so would be interesting to see any new methods.
All the best, and thanks for the great reads you’ve given me down the years…
Hi Marc,
It’s good to meet you. Thanks for being a long time reader here. I’m going to be including keyword research and context vectors in my presentation for Pubcon and will be publishing my presentation for that along with a blog post. So, I will be covering it in more depth. But ideally, when you are optimizing a page for a particular word that might have more than one meaning, you want to look in knowledge bases such as Wikipedia or wikidata, and see if there are additional words from that source that you can add to your page that helps give it more context. My presentation expands that into using topical modeling, so in addition to adding words from a knowledge base, perform a search for that keyword on Google, and find the top ranking pages that share the same meaning, and look for phrases that tend to co-occur on those pages, and consider adding those to your page that you are trying to rank for the same keyword. Doing so helps you let Google know what the meaning behind your page is better. The idea is to better match your pages up with meanings or concepts.
Hi Bill,
Whether Google is changing, but there are still people I know who are using old way to get rank. So, we must change in keeping the factors in mind.
Hi Robin,
We’ve been seeing the SEO industry going through changes for as long as people have been promoting websites. I think it is helpful to keep an eye open for changes, and experiment and try new things to keep up with changes. For instance, people who haven’t been trying to use structured data and json-ld, and building knowledge panels for clients should ideally be trying to do so. Same with featured snippets and rich snippets. Addressing different meanings of keywords by looking at knowledge bases for related words and concepts is worth the effort; Google is showing us that it is concerned with those things.
Huh, if you think about it, I’m pretty sure this is actually great news for SEOs. It’d seem that it means that there will more or less be multiple different SERPs for each “version” of a word (ie. “rock”) and thus, less competition for that particular keyword.
In other words you should find it easier to rank for rock (music) now that you won’t have to directly compete with pages about rocks.
Google changes so much every single day… I am already very scared of the infinite knowledge..anyways
Really good article , truely appreciated your work
keep up the good work and keep inspiring us…
Hi Bill,
Your ideas are interesting and seem to make sense. The future is still hidden and no one definitely knows what will with SEO and Google
Great post! I am actually getting ready to across this information, is very helpful my friend. Also great blog here with all of the valuable information you have. Keep up the good work you are doing here.Well, got a good knowledge.
Hii Bill,
This is always something new to learn everywhere, Google does changes almost every day.So, we must change in keep in mind all google factors for get a high rank in google .
Google changing it so much day by day to become more correct by improving there artificial intelligence and i think its a good change of topical search on context based results. Looking forward to appear it in search results.
Hello Bill,
Great tips over here 🙂
Indeed Google changes every day, its search algorithm to their every single unit.
It sometimes takes a lot to rank our web sites among the SEO, so that we could hall have a win win situation for our blogs.
Keyword research is always something that I have struggled with and as you mention about the Content vectors, I am looking forward to
grab my finger upon this.
Thanks for the amazing share.
Shantanu.
Great post! I am getting ready to across this information, is very helpful and useful . Also great blog here with all of the valuable information you have given. Keep up the good work and ensure the success of all budding digital marketers you are doing here.Well, got a good knowledge.
Thanks Bill, awesome article. Love the emphasis on “context vectors”, I think others should take the time to learn about this with specific keywords, and really think about what people are searching for when typing it into Google. Amazes me the changes Google continues to make every day. In your opinion, what’s the most crucial thing about topical search results that need to be considered? Additional keywords on top of “horse” or potentially quick filter topic buttons?
Hi Andy,
I’ve been seeing Google place a lot of emphasis on authority of search results they might rank highly. If you can create pages that show off expertise, authoritativeness and trust (as described in the Google rater’s guidelines), and if you can show off the right context that a searcher may be looking for. that should make a difference.
Search Qualith evaluators guidelines
I see this to be a KEY to the best answer to question…
Authoritativeness and trust (as described in the Google rater’s guidelines), should help to make a difference.
Hope this helps…
Hi David,
I think there is a lot of value in that rater’s guidelines
Just Awesome post boss.
Actually I’m looking for that valuable information & I’m glad to see your informative post here it’s really helpful & useful for us and yeah I’m gonna share your wonderful information with my friends, thanks for sharing & thanks alot for ur big support great job boss… Keep Posting
Love You
Tc ðŸ˜Å
HI
Your ideas are interesting and seem to make sense.Google does changes almost every day. we must keep in mind all google factors for get a high rank in google .
Thanks for sharing Such great post with Us.
Came back to this blog after ages and found again fresh and extremely informative content specially about Google search engine. Surprising to know how Google found the categories and context of the items searched.
This website is very informative to read. I am a huge follower of the things you talk about. I also love reading the comments, but it seems like a great deal of readers need to stay on topic to try and add new things in the original topic. It and also added your RSS feeds, so when I have time I will be back to read a great deal more.
This is definitely the biggest marketing resource list I have ever seen online. You have spent quite a good time to research about all those resources to help people learn from basics to advance levels of marketing and how they are going to work with competition and improve their marketing
At this moment in time SEO is very unpredictable. Artificial intelligence will change so many different aspects of how we search and how search engines gives us the information we desire. Things are about to get even more complex. Oh, I almost forgot to mention Voice search as Alexa and Google Home invade our homes. Surely paid marketing is a sure way forward in 2018? At least until things calm down a little.
Very nice blog on this website. It is really difficult to get this kind of with useful
information. I am relieved I came upon this site. I will eagerly look forward to your
upcoming updates .
A fragrance retailer once mentioned to me in relation to search volume analysis for Cologne (also a city in Germany) whether I could ascertain what percentage was regarding their market and what was in relation to the city. I thought it was quite a good question! Will be interesting to see how Google’s algorithms develop into the future.
Thanks for the post!
The recent algorithm hit most of us hard and now clueless and what to do next. But will look into new methodologies and sites like this. Thanks