is how search engines are learning how to interpret what a user on a search engine is intending with their search.
It is a form of text analysis to enable others computers to understand the content within a webpage.
is similar to
except instead of just repeating keywords over and over within the same website, it is repeating themes from keywords.
Moz.com gives an example of entity salience would be if a website references Iron Man, Tony Stark, Pepper Potts and Science FictionMoz seo concepts. Even if the phrase Marvel Comics only appears once it will have a strong correlation with the entity salience of Marvel Comics, because it is so heavy in the topic of keywords that often fall within other realms of Marvel Comics. Another example is if you have the terms: dinosaurs, t-Rex, and Michael Crichton the entity Jurassic Park will be saturated with keywords.
is just one of the
that can be used to strength the overall
. This is a sign of the shift in search engines from just matching specific keywords to the body of websites to trying to understand the root query of a user and answer that first. The more words that are related to an entity, the higher the
Entity salience is a marrying of
sentiment analysis (
the machine learning process of understanding the sentiment behind a given amount of text) and Search Engine Optimization. So far there isn’t a strong competitor in giving search optimization tips or evaluating a salient entity. There are companies built around
in the business to business realm that operate outside of
. One prominent company is Lexalytics. They primarily do a process they call Lexanalytics sentiment analysis and take other business’ social media data or reviews and help them understand the sentiment behind it.
This form of SEO is makes it difficult to do bad
. One reason is because Entity Salience is a bit of a black box on how Google, or Bing uses it to evaluate a website. There are many different algorithms that can be run to evaluate sentiment analysis. None are perfect so this is just one of the tools in the SEO toolkit.
However, since it is hard to mimic it by illegitimate websites it is very powerful and it’ll do wonders to improve your webpage ranking. It’s important to keep a natural flow of content. Even though this is a strong competitor in page rankings it also is computationally more difficult than keyword pumping into a webpage post.
Google gives a window into how it researches entity salience Google Salient entity research. There they state:
“Although many NLP systems are moving toward entity-based processing, most still identify important phrases using classical keyword-based approaches. To bridge this gap, we introduce the task of entity salience: assigning a relevance score to each entity in a document. We demonstrate how a labeled corpus for the task can be automatically generated from a corpus of documents and accompanying abstracts. We then show how a classifier with features derived from a standard NLP pipeline outperforms a strong baseline by 34%.”
This is the wave of the future in search engines. As computers try to get smarter at understanding what we mean there will come a time when we will get frustrated with how smart they are trying to become. Currently there is a small backlash of how often Google removes keywords from a user’s search. Most users are unaware of how to force google to include all keywords. But from that example Google is trying to understand our questions, and our websites, better than we can articulate them.