When to use Machine Learning Models in SEO and Which ones to use - Lazarina Stoy for inOrbit 2024
•
0 likes•14 views
This talk is a walk-through of different ways you can incorporate machine learning in SEO tasks. It will involve a speed-run of different task categories/ aspects of SEO work, and the models that you can use for this purpose, and the results of a comparative analysis of how they perform. Listeners will leave with (1) understanding of what are different ML models, and where to incorporate them in their day-to-day SEO work, (2) why and how to choose one solution over the other, (3) how to get started with the recommended ones (will be sharing videos/templates/walk-throughs) This session focuses on: The process of incorporating ML models and the aspects of SEO work when they can be incorporated (e.g. image captioning, generative work in content or meta elements, content localization, etc) A summary of comparative analysis work I’ve done where I’m comparing the performance of different models for specific tasks in SEO, and providing a recommendation of which one to use for what task and why Summary of steps and costs, plus templates/code to use Who is this talk for? Any SEO (agency), SEO manager (In-house), or site owner Team Leads, looking to upskill their teams and processes to rely a bit more on automation People interested in automation and ML/AI, and interested in going beyond chatGPT Those interested in saving some time in their day-to-day tasks via automation
Recommended
Recommended
More Related Content
Similar to When to use Machine Learning Models in SEO and Which ones to use - Lazarina Stoy for inOrbit 2024
Similar to When to use Machine Learning Models in SEO and Which ones to use - Lazarina Stoy for inOrbit 2024 (20)
Recently uploaded
Recently uploaded (20)
When to use Machine Learning Models in SEO and Which ones to use - Lazarina Stoy for inOrbit 2024
- 1. @lazarinastoy | @mlforseo LazarinaStoy.Com When to use Machine Learning Models in SEO and which ones to use
- 8. How well do you know what your content is about?
- 9. clustering and classification let’s talk about
- 10. @lazarinastoy | @mlforseo LazarinaStoy.Com Classification sorts data into specific categories using a labeled dataset.
- 11. @lazarinastoy | @mlforseo LazarinaStoy.Com Clustering is partitioning an unlabeled dataset into groups of similar objects.
- 12. @lazarinastoy | @mlforseo LazarinaStoy.Com Clustering is partitioning an unlabeled dataset into groups of similar objects. We’ll go back to this in a moment.
- 13. How important is classification, really?
- 15. @lazarinastoy | @mlforseo LazarinaStoy.Com With Google’s Natural Language API, you can classify documents in 1,300+ predefined categories
- 16. @lazarinastoy | @mlforseo LazarinaStoy.Com Process will take no more than 20 minutes
- 17. @lazarinastoy | @mlforseo LazarinaStoy.Com Enter your URLs and content
- 18. @lazarinastoy | @mlforseo LazarinaStoy.Com Run the script via a formula to get classification label & confidence
- 19. Plug-and-play template in Looker Studio
- 23. @lazarinastoy | @mlforseo LazarinaStoy.Com With Open AI’s GPT4 or ChatGPT, results are a hit or miss.
- 24. GPT-4 vs. Google Cloud: Performance comparison on 9 SEO tasks
- 25. Do you need content clustering?
- 27. @lazarinastoy | @mlforseo LazarinaStoy.Com LDA is great (even though it’s a bit old)
- 28. Introduction to Probabilistic Topic Models (Blei, 2012)
- 29. Introduction to Probabilistic Topic Models (Blei, 2012)
- 30. @lazarinastoy | @mlforseo LazarinaStoy.Com LDA emerged to: ● remove dependency on links by introducing the “things” concept and topic/term understanding ● enable computational understanding of topics and terms and their importance ● highlight that each page will have multiple different topics or subtopics addressed, which might be of value to different people and should be understood and surfaced in results
- 31. @lazarinastoy | @mlforseo LazarinaStoy.Com WATCH THE DETAILS LATER I’ve recorded a step-by-step tutorial on doing topic modelling using a no-code, publicly-available, web-based app using LDA.
- 32. @lazarinastoy | @mlforseo LazarinaStoy.Com Identify topics CRAWL Export the content from the website UPLOAD + FINETUNE Upload the files to the web app + finetune DOWNLOAD Download all files EXPLORE + BUILD Explore the outcome and build your deliverable
- 33. @lazarinastoy | @mlforseo LazarinaStoy.Com Identify topics CRAWL Export the content from the website UPLOAD + FINETUNE Upload the files to the web app + finetune DOWNLOAD Download all files EXPLORE + BUILD Explore the outcome and build your deliverable Process will take no more than 30 minutes
- 35. @lazarinastoy | @mlforseo LazarinaStoy.Com Topic models Page Info
- 37. BERTopic is another great alternative
- 42. @lazarinastoy | @mlforseo LazarinaStoy.Com Categorise/ discover patterns and topics on site content → Identify opportunities for internal linking → identify what your site is about and whether it aligns with business positioning → Identify the topics that your competitor site tackles
- 43. @lazarinastoy | @mlforseo LazarinaStoy.Com Categorise/ discover patterns and topics on YouTube titles or video catalogs → Quickly understand competitive landscape in hundreds or thousands of videos in a niche
- 44. @lazarinastoy | @mlforseo LazarinaStoy.Com Categorise/ discover patterns and topics in first-party data (any kind of user forms) → Quickly see what topics your feedback is centred upon Bonus points: Tie this analysis with sentiment analysis.
- 45. Do you really understand what makes up a topic?
- 46. keyword clustering let’s talk about your fancy software, and how to replace them
- 48. @lazarinastoy | @mlforseo LazarinaStoy.Com Keyword clustering is sub-topic keyword extraction in text-based documents.
- 55. @lazarinastoy | @mlforseo LazarinaStoy.Com Data to analyze ● your content ● competitors’ content ● your SERP data versus competitors’ ● first-party data ● UGC ● social mentions
- 56. @lazarinastoy | @mlforseo LazarinaStoy.Com With Google’s Natural Language API, you can analyze entities, and entity sentiment.
- 57. @lazarinastoy | @mlforseo LazarinaStoy.Com Process will take no more than 20 minutes
- 58. @lazarinastoy | @mlforseo LazarinaStoy.Com Run the script via the menu to extract entities from content
- 59. Get entity data in seconds
- 60. @lazarinastoy | @mlforseo LazarinaStoy.Com Map to topics, labels or intent classifications
- 63. Does it matter how you do it?
- 64. @lazarinastoy | @mlforseo LazarinaStoy.Com Yes, use a tailored model. 53 16
- 65. GPT-4 vs. Google Cloud: Performance comparison on 9 SEO tasks
- 66. @lazarinastoy | @mlforseo LazarinaStoy.Com Entity analysis work is central to multiple SEO projects (so, pretty important to get it right)
- 72. Do you need to understand text structures, at scale?
- 73. let’s talk about fuzzy matching
- 74. @lazarinastoy | @mlforseo LazarinaStoy.Com A quick and dirty way for calculating similarity between two strings
- 75. @lazarinastoy | @mlforseo LazarinaStoy.Com WATCH THE DETAILS LATER I’ve recorded a step-by-step tutorial on using fuzzy matching for things like: ● Identifying link opportunities ● String Similarity Analysis ● redirect mapping of URLs I
- 76. @lazarinastoy | @mlforseo LazarinaStoy.Com Assesssimilarity CRAWL + MODEL Export the content from the website ORGANIZE Теxt to analyze - titles, URLs, hreflang tags… ASSESS Human-check similarity scores. ACT Implement or recommend
- 78. @lazarinastoy | @mlforseo LazarinaStoy.Com Language use analysis can also be useful for scaling generation of or identifying opportunities for Structured Data (schema) mark-up
- 80. @lazarinastoy | @mlforseo LazarinaStoy.Com From blog post Run the cell, paste the content Download the output JSON
- 81. @lazarinastoy | @mlforseo LazarinaStoy.Com → Script tokenizes the text, discovers the questions, and pulls the answers
- 82. @lazarinastoy | @mlforseo LazarinaStoy.Com → Script organises these into a schema dictionary, which is saved as a JSON file
- 83. @lazarinastoy | @mlforseo LazarinaStoy.Com What took minutes even with the most straightforward methods (e.g. using a schema markup generator, and copy-pasting individual questions into it), took seconds.
- 84. Are recent updates making you rethink content distribution?
- 85. Content Transformation let’s talk about something a bit more advanced (but still beginner-friendly)
- 86. @lazarinastoy | @mlforseo LazarinaStoy.Com Improve brand omni-presence and content accessibility Both users and search engines want to see multi-modal presence for high-value sites. Meaning: ● Text to video ● Videos to text ● Text to audio ● Audio to text ● Text summaries for content distribution on social media
- 87. @lazarinastoy | @mlforseo LazarinaStoy.Com Needless to say scaling production here is pretty significant for organic growth.
- 88. @lazarinastoy | @mlforseo LazarinaStoy.Com You have a library of videos on YouTube but no blog? → Scale it’s launch quickly through transcription.
- 89. @lazarinastoy | @mlforseo LazarinaStoy.Com You have a library of high-performing blog posts but no presence on YouTube/TikTok? →Scale production with text to speech.
- 90. @lazarinastoy | @mlforseo LazarinaStoy.Com You have a library of high-performing blog posts but no content distribution? →Transform blog posts to insightful posts for social media.
- 92. half of the people in the room are thinking
- 93. trafficthinktank.com “ I got this.”
- 95. mindset + community + resources