From Knowledge Graphs to AI-powered SEO: Using taxonomies, schemas and knowledge graphs to improve search engine rankings and web publishing workflows

Workshop Room 1

Do you want to learn how to use the low-hanging fruit of knowledge graphs — schema.org and JSON-LD — to annotate content and improve your SEO with semantics and entities?¬†This hands-on workshop with one of the leading Semantic SEO practitioners will help you get started.

This workshop is kindly sponsored by Wordlift – The Artificial Intelligence you need to grow your audience. Wordlift is a WordPress Plugin that does what an SEO expert would do.

Key Topics

SEO, Structured Data, Knowledge Graphs, OpenRefine

Target Audience

The tutorial will be of high value to marketeers with a data-centric approach, content creators and data practitioners who are (expected to be) involved in developing and/or enriching knowledge graphs to markup online content, to improve SEO (Marketeers, SEOs, Content Creators, Information Architects, Data Modelers, etc) and to grow online businesses.

Level

Intermediate

Prerequisite Knowledge

The tutorial will be self-contained but participants should have some prior knowledge of linked data, structured data markup and the schema or vocabulary.

Hardware & Software Requirements

A modern web browser

What you’ll learn

After this tutorial participants will be more aware of structured / linked data and Semantic SEO. They will be equipped with concrete strategies and techniques to leverage on public graphs such as Wikidata and DBpedia for their own projects. They will also learn how existing data – within their organization – can be used in semantic markup, SEO and web analytics.

Description

Ever since schema.org was announced and search engines transitioned from being “tools that help us find relevant content” into “cognitive applications that are capable of answering questions”, the way that content is conceived, organized and promoted online has changed significantly. Practices in the digital marketing space and more specifically in the SEO industry have evolved accordingly.

Google had a massive impact in making the content industry aware of how semantics, linked data and knowledge graphs can inform the user experience. From Google Search to Google Discover, from Alexa to the Google Assistant data is powering end-to-end the interactions in a new AI-first digital environment.

While several large organizations, including Airbnb, Amazon, Linkedln, and Zalando, had invested in creating their own knowledge graphs for data integration, data analytics, semanbc search and other applications, WordLift’s mission has always been to use linked knowledge for promoting and organizing web content (or more simply to improve SEO and how content can be found by search engines, personal digital assistants and content discovery platforms).

This tutorial will take participants into a deep dive in Semantic SEO and teach them how to build linked data that search engines can use to properly understand the content of websites. More importantly, it will provide them with concrete strategies on how this same data can also be used internally for improving the understanding of the traffic on their website and/or for other use cases (i.e. chatbots, content recommendations, …).

Session outline:

The tutorial will consist of two parts. The first part will be lecture-based and will ensure that all participants share some common understanding and mindset about Semantic SEQ and structured data. This is important as practitioners might have different backgrounds and different roles (Marketeers, SEOs, Content Creators, Information Architects, Data Scientists, etc)..

The second part will be highly collaborative and interactive. The participants will form small teams, each of which will choose a reference website and will start with a “semantic audit” using python (code will be made available in iPython). Each team will then extract a selection of the most common named entities on the website that will be enriched using OpenRefine with queries back and forth against DBpedia and Wikidata. Finally, one of the resulting datasets will be injected in the form of JSON-LD to enrich the content of a website.

Sample schedule:

* 9:30 – 9:40 – Introductions (10 min)

Participants introduction
Goals and scope of the tutorial

* 9:40 – 10:00 – Semantic SEO and Structured Linked Data (50 min)

Elements of Modern SEO and how machine learning and linked data are used in Google Images, Google Assistant and Google Discover.
Why structured / linked data is important
What to expect from marking up your website
How to run a Semantic Audit of a website

* 10:30 – 10:45 – Break (15 min)

* 10:45 – 12:15 – From entities and graphs to JSON-LD (90 min)

Extracting the most relevant entities of a website
Enriching data with the help of DBpedia and Wikidata using OpenRefine
Figuring how to re-use the data to markup the pages
Evaluating how to re-use the data for web analytics and content recommendations

* 12:15 – 12:30 – Conclusion

Take-Aways
Additional resources


Workshop