From Knowledge Graphs to AI-powered SEO – The Practice | Andrea Volpini | Knowledge Connexions 2020

The practice of using schemas, knowledge graphs and NLP to develop a long-tail SEO strategy With advancements in Artificial Intelligence happening everyday and Google pushing search results into the realm of conversational user interfaces, web publishers can benefit from deep learning (DL), structured linked data and natural language processing (NLP) Building the knowledge graph using schema.org is foundational and enables organizations to exploit their data in modern SEO and digital marketing. Adding semantic markup helps search engines understand the content we write and helps us re-organize content to fit the needs of our audience In this workshop, we will learn how to use knowledge graphs to discover new search-demand areas and build dynamic pages that can target long-tail queries. Long-tail are “unpopular” (i.e., low volume) and highly-focused search queries that tend to convert exceptionally well After this workshop participants will be more aware of structured linked data and Semantic SEO. They will be equipped with concrete strategies and techniques to leverage on existing data - within their organization - for improving their publishing workflow and for discovering new long-tail queries. We will also cover some essential elements of natural language generation using Google’s T5 Text-to-Text Transfer Transformer Model Key Topics The interplay between AI and content editors The role of the CMS Challenges of using State of the Art AI language models Costs of training Working with long-form content and the advantages of distilled models Analyzing Google’s SERP using your Knowledge Graph Target Audience Web Publishers Marketers with a data-centric approach Content creators and data practitioners who are (expected to be) involved in developing and/or enriching knowledge graphs to improve SEO and to grow online businesses Marketeers SEOs Content Creators Information Architects Data Modelers Goals Get hands-on experience using schemas, knowledge graphs and NLP to develop a long-tail SEO strategy Session outline From queries to entities to pages Extracting the most relevant entities behind a search query Training a language model for content summarization Building dynamic pages using content in your Knowledge Graph Evaluating how to add relevant content recommendations and FAQs Conclusion Take-Aways Additional resources Format This class is highly collaborative and interactive. Participants form small teams, each of which will work on the reference website and run a "search intent investigation" using Python (code will be made available in Google Colab) and the reference website's Knowledge Graph. Each team will use NLP to highlight new untapped search opportunities. Finally, we will see how to reverse-engineer the results of Google Search for a given query and how to assemble a web page in WordPress that can attract traffic. The web page will include machine-generated content created using Google's T5 Text-to-Text Transfer Transformer Model. Level Intermediate - Advanced Prerequisite Knowledge From Knowledge Graphs to AI-powered SEO - The Theory Google Colab and WordPress will be used -- Andrea Volpini: CEO, Wordlift Andrea Volpini, CEO of WordLift, is a visionary entrepreneur, now focusing on semantic web and artificial intelligence. Andrea works at the intersection of the semantic web, AI, and SEO, helping brands worldwide increase their organic search visibility, traffic, and conversions. Andrea has 20+ years of world-class experience in digital marketing. Previously, he was a cofounder of InSideOut10 and kick-started Redlink, a research spin-off focusing on AI and information extraction.