This is a guide for machine learning for beginners, tailored to the SEO industry, aimed at breaking down the challenges that hold us back from experimenting, the breakdown of machine learning's main characteristics to help us understand how to implement it a bit better, and the ways we can embed advanced technology into our daily practice.
30. What to do
Waiting To
Get Started?
Search machine
learning in 10
minutes. Follow
along.
Start small but do
something today.
Awaiting
Perfect
Conditions?
Struggling or
Tried and
Failed?
Build a habit and
track your
progress.
Start small and
remain consistent.
cut scope or
change direction.
Start small to get
back into it.
@lazarinastoy | #WTSFest
31. What to do
Waiting To
Get Started?
Search machine
learning in 10
minutes. Follow
along.
Start small but
something today.
Awaiting
Perfect
Conditions?
Struggling or
Tried and
Failed?
Build a habit and
track your
progress.
Start small and
remain consistent.
cut scope or
change direction.
Start small to get
back into it.
@lazarinastoy | #WTSFest
44. Aim to drive value
and have fun with
the process.
45. @lazarinastoy | #WTSFest
Knowing
● what model to use
● how to find and implement it
quickly
● how to drive value via ML
is the perfect way to
start.
50. You have
labelled data
to validate
results
You don’t
have a way
to validate
results
Find patterns and group
based on similarity
Simplify or transform your
data
Make predictions
Split into groups, based
on existing classes
@lazarinastoy | #WTSFest
65. Find a model or procedure that
makes best use of historical
data comprised of inputs and
outputs in order to skillfully
predict outputs given new and
unseen inputs in the future.
-Jason Brownlee
”
66. A model or procedure that
automatically creates the most
likely approximation of the
unknown underlying
relationship between inputs and
associated outputs in historical
data.
-Jason Brownlee
”
68. 1. Writing Meta
Descriptions
● Input data? Textual (page content).
● Supervised or unsupervised? Unsupervised
○ It is transformational (Page content to Page Summary
in less than 160 characters)
○ It can also be generative (write them from scratch)
● Is it mission critical? No.
● Different results okay? Yes.
● Explanation of process needed? Not really.
● Outperforms average methods? Yes, much faster.
71. 2.Title / H1
Optimisations
● Input data? Textual (page content).
● Supervised or unsupervised? Unsupervised
○ It is transformational (Page content to Page Summary
in less than 60 characters)
○ It can also be generative (write them from scratch)
● Is it mission critical? Hm, debatable (ask HMRC 👀)
● Different results okay? Again, debatable.
● Explanation of process needed? Kind of.
● Outperforms average methods? Yes, much faster. Not
better.
73. 3. Image Alt tag
Generation
● Input data? Image
● Supervised or unsupervised? Supervised
● Generative
● Is it mission critical? No
● Different results okay? Yes.
● Explanation of process needed? Not really.
● Outperforms average methods? Yes, much faster.
75. But might also
involve…
● Predicting traffic / revenue based on
presence of keyword in the title/ h1 to get buy
in on proposed changes
● Updating internal links
● Researching keywords for new content
updates
● FAQ schema implementation
85. get ready to practice
1 2
3 4
Install Anaconda
and pip. Set up
Colab.
Install the
main ML
libraries
Then get on
with your role
as per usual
Get your foot
through the
door with some
daily practice
@lazarinastoy | #WTSFest
88. Deconstruct each task you
encounter
Assess whether ML is the correct
approach to solve the problem
Adjust (prepare) or gather your
data (if needed)
Build or test scripts,
libraries and tools.
Assess results.
Scrutinise the output.
Compare, if possible.
Document the journey.
Build your deliverable.
@lazarinastoy | #WTSFest
101. Test, test, test.
Test scripts every day. Dissect,
analyse, understand them.
Find what you like and dislike
about them.
Note best practices.
Building will become easier
after.
102.
103. CREDITS: This presentation template was created by Slidesgo, including icons
by Flaticon, infographics & images by Freepik and illustrations by Storyset
Read the talk re-cap & drop any Qs
in the comments 🔥:
lazarinastoy.com/beginners-guide
-to-machine-learning-for-seos/
104. CREDITS: This presentation template was created by Slidesgo, including icons
by Flaticon, infographics & images by Freepik and illustrations by Storyset
Check out the ML for SEOs collection
on GitHub (scripts, notebooks, and
tools):
github.com/lazarinastoy/beginners-
guide-to-machine-learning-for-seos