Posted by Emilia Gjorgjevska 2 months ago

Table of content

Understanding Information Extraction

Search engines have become part of our daily lives. We use Google, Bing, Yandex, Baidu, DuckDuckGo, etc. as the main gateway to find information on the Web. We use Facebook or LinkedIn to search for people, associations, and events. We rely on Amazon or eBay for product information and comparisons, while when it comes to music we like to play stuff on YouTube or Spotify. We are decreasingly reliant on apps’ features to find connections, dispatches, notes, timetable entries, etc. We’ve grown habituated to anticipating a search box nearly near the top of the screen. We also increased our expectations to get fast responses back while we search for things.

This is how we and machines developed the need to connect information needs with information concepts and precisely how the fields of information retrieval and information emerged. Search queries are expressions of our information needs while the ranked lists of information objects are the answers provided back to the searcher.

In the past, while search engines were in the early development phase, it was easy to play with keywords and links and use shady techniques to easily rank on search engines. Now, with the development of advanced technologies and algorithms like Multitask Unified Model (MUM) and Locked Image Tuning (LiT) we need to embrace more intelligent solutions and carefully crafted content to answer user queries in the best possible way and provide a satisfactory user experience.

When dealing with natural language queries, we need to distinguish between named entities and concepts.

Named entities are real-world objects and they can include:

  • persons like Martin Splitt, John Mueller, Sundar Pichai;
  • locations like Mountain View, Silicon Valey;
  • organizations like Google, Pinterest;
  • products like Google Assistant, and Google Cloud;
  • events like Search UnConference, Knowledge Graph Conference, etc.

Concepts are the opposite of named entities and they represent abstract objects. Some examples include:

  • mathematical and philosophical concepts like distance, axiom, quantity;
  • physical concepts or natural phenomena like gravity, force, and wind;
  • psychological concepts like emotion, thought, and identity;
  • Social concepts like authority, human rights, and peace.

Named Entity Linking (NEL) And SEO – Smart Search Performance Optimization

In many information extraction applications, entity linking (EL) has emerged as a crucial task in understanding named entities through their linked descriptions obtained from a knowledge base like YAGO, Wikidata, DBPedia, and similar. This process is better to know as semantic mapping or semantic linking in the computer science world:

The first step of the entity linking process is mention detection where we need to obtain the list of named entities in the text. A named entity compared to a casual entity is an entity that is already defined in a knowledge base or an NLP model;

The second step is candidate entity ranking where after analyzing the user query we obtain several entity candidates in a ranked order. E.g. depending on the context, the entity [apple] can refer to the fruit or Steve Jobs’s company;

Entity interpretation is the final step where we decide on the best candidate for a given query input from the user and use this candidate to retrieve more information from a knowledge base back.

Entity linking can boost your SEO performance by improving:

  1. Your mobile results: entities help improve mobile capabilities and mobile-first indexing which became dominant in search and it’s growing every year
  2. Natural language and image understanding for rich snippets: things like photos, customer ratings, and product reviews belong in this group;
  3. Translation optimization: synonyms, homonyms, context clues and query facets, and entity disambiguation help in translation improvements;
  4. Increased traffic and conversions: entity linking and entity disambiguation help search algorithms understand your content better and distribute it to more targeted users so that you’ll have more visits back and increased chances to convert them into customers.

Entity-Oriented SEO Is The Future Of The Search

Entity-oriented search and optimization give context to your website. That is why it’s important to work with entities because they help to connect the world’s information together and therefore get relevant results back when searching.

One effective way to do so is to use a specific type of schema markup that will contain all the entities in the ABOUT and MENTIONS schema attributes. Another, more advanced way is to create a knowledge graph out of your content and publish it as linked open data on the Linked Open Cloud. 

This is exactly what we do with WordLift. We make a shift from the typical link-building mindset and keyword-oriented search to entity-oriented search and a more advanced link-building approach that employs entity linking between the entities and their respective descriptions in popular knowledge bases. We are also a proud member of the DBpedia Association and actively contribute to the growth of Wikidata. Wordlift is publishing high-quality web-scale knowledge by following Tim Berners Lee’s 5-star principles from CMS and the e-commerce platform. This means also building links with other public graphs.

One thing is clear: user needs are constantly evolving and it is becoming harder to keep up. Do stuff that matters and make use of linked open data to stay relevant.

Must Read Content

Why Do We Need Knowledge Graphs?
Learn what a knowledge graph brings to SEO with Teodora Petkova

2x Your Video SEO in 3 Simple Steps
Use videos to increase traffic to your websites

SEO Automation in 2021
Improve the SEO of your website through Artificial Intelligence

The Power of Product Knowledge Graph for E-commerce
Dive deep into the power of data for e-commerce

Touch your SEO: Introducing Physical SEO
Connect a physical product to the ecosystem of data on the web

Are you ready for the new SEO?
Try WordLift now!

Learn how to use structured data to get better SEO rankings. Practical SEO tips, webinars, and case studies right in your inbox once a week.

Learn how to use structured data to get better SEO rankings. Practical SEO tips, webinars, and case studies right in your inbox once a week.

Congrats you are in!

Stand out on search in 2019. Get 50% off WordLift until January 7th Buy Now!

x