The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Share
π§ Knowledge (Graphs) In The AI Era
Published over 1 year agoΒ β’Β 6 min read
β
Use Data Or Be Used By Data!
The May 26 issue of Seotistics is here for you!
Knowledge graphs have been making their way through in the last few years.
Small and medium companies have no clue but corporations are currently employing such technologies!
So why am I even mentioning them?
And what have these strange things to do with AI?
P.S. This topic is huge, expect a more proper coverage in either my website or course.
Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this tutorial to do it.
Thanks to my dear readers and supporters for sustaining my growth.
We can definitely do more though, I am investing more efforts into our Discord community now.
Don't hesitate to speak and interact, this is a period of transition and new experiences are necessary!
To come next:
"Analytics for SEO" ebook v9
More additions to "Analytics for SEO" course
More content for "Think Like a Web Analyst"
Freebies (one already out)
Articles
The Google Flashback
In the far May 16 2012, Google mentioned to have a Knowledge Graph.
In the previous seasons...
Back then, it didnβt really matter for practitioners.
Some concepts were introduced here and there and tools like InLinks and WordLift managed to conquer some market share in the SEO world.
We are not here to talk about one channel, though.
What The Heck is a Knowledge Graph (KG)
A knowledge graphs is a type of graph, which is a set of nodes and edges.
Nodes = entities or concepts, e.g. Web Analytics
Edges = relationships, aka connections between nodes.
But the entire idea revolves around the concept of an RDF triple (subject, predicate, object).
E.g. Paul (SBJ) eats (PRD) an apple (OBJ).
In this case, there is a directed edge, Paul is pointing to Apple.
This is a powerful concept because you can classify the concepts that belong to a given domain in a company.
You can say KGs describe entities and their relationships.
If you strip the data out a KG, you get an ontology.
An ontology doesn't go into detail but it defines how these relationships are handled and which classes you have in a domain.
Some of these concepts I will define in a separate issue/article π.
That's Not All!
What does it even mean?
It means that you can explain better what is what and how the business works.
Letβs consider the following simplified example:
β
Now, the cool part about this is about the properties the graphs have:
Connecting different domains aka bypassing silos, e.g. product to sales
Contextual Analysis
Inference
We will see some examples soon π
You can also use a language like SPARQL to extract data from a KG.
Imagine SQL but for RDF triples, this is the gist!
I studied a little bit of it and was super confused by some concepts back thenβ¦ now itβs much more clear.
Oh, and did I mention we have LLMs to help us too?
Displaying cat names from Wikidata
Services like Wikidata stored RDF triples and offer spaces like this to test your SPARQL skills.
Or just use good ol' Python run queries from your notebooks:
β
P.S. You can also play with KGs in Neo4j but I personally dislike their UI.
Taxonomies
These are used to rank and classify metrics, for example.
Say you want to have your users metrics together, how would you classify them?
For large projects, these are a must because they allow you to define metrics and have a clear understanding of what is what.
Also, in big organizations there is no alignment between departments, itβs like a set of micro-states fighting each other.
A sample taxonomy that is classifying product categories in the Ecommerce category inside Web data.
On a recent project of mine, I had to create a taxonomy for some customer metrics and unify their definitions.
It's extremely invaluable to have clarity and an idea of what you should consider for the business.
The Intersection With Analytics
We are used to tables and β2Dβ analysis.
Say you want to understand how many people from the USA who bought from you given a campaign happened last month coming from social media.
If you have some experience with tools, this shouldnβt be super hard to pull off.
But what I asked you to tell:
"which customers who bought products similar to Product A, were influenced by campaigns targeting similar demographics, and have purchasing patterns similar to our high-value customers?"
This is where a KG can do some inference and figure out relationships you would miss.
The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Read more from Seotistics - Web Analytics + Business + Strategy
Use Data Or Be Used By Data! The October 19 issue of Seotistics is here for you! I am just back from Rome and AI didn't sleep, we are full of news and updates. As you know if you are working with automation, you need to know this stuff! As I try to recover from multiple conferences in a row, here is the latest summary of what's going on. Most of these topics will be expanded in the next issues as they are too new π Please move this email to your Primary inbox or reply to it. This is to...
Use Data Or Be Used By Data! The October 12 issue of Seotistics is here for you! Measurement did change in the last year and I am here to remind you some basic facts. AI is for sure one of the main culprits but not the only one. I am also writing a huge article on LLM measurement, stay tuned π Hectic week, too much to do, including course updates. Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this...
Use Data Or Be Used By Data! The October 5 issue of Seotistics is here for you! The last time we saw some nice workflows you can use in real time for your small website. Well, I show you what else you can do and the latest news in the Web Analytics world :3 Nowadays you can have a better Analytics stack than your competitors just with common sense. There are big changes coming ahead and 2027 will be THE year. P.S. Go check my previous issues if needed, this is a sequel. Please move this email...
