When it comes to machine learning and SEO, there have been a number of advancements in the past decade with Google getting a lot of publicity and praise for projects such as RankBrain, BERT, and SMITH.
That being said, they are not the only search engine in the world making great strides in advancing machine learning.
Over a similar time frame to Google, Yandex have released similar projects into their ranking processes such as MatrixNet, Palekh, and it’s second more refined iteration Korolyov, and most recently YATI. Baidu have also been involved in developing machine learning technologies for search, with their more prominent ML model being ERNIE.
During this article, I’m doing to use the word Transformer a fair few times, and I think it’s important to have a top-line understanding of what a Transformer is, and how models like BERT and SMITH, and connected to YATI and ERNIE.
What Are Transformers?
In simple terms, a transformer is a deep-learning model that is used in recurrent neural networks (RNNs) for handling tasks involving sequential data and natural language.
They facilitate something known as parallelization, meaning that input data doesn’t need to be processed in order, making it possible to process and facilitate much larger, greater scale data-sets.
What Is YATI (Yandex)?
Since 2017 there has been little in terms of new ML technology from Yandex, but at at the end of 2020 Yandex launched YATI, a new ranking algorithm based on transforming neural networks.
YATI stands for Yet Another Transformer with Improvements.
It may not be poetic, but it’s been hailed as the most significant and impactful change that Yandex has made to it’s search ranking algorithms since the introduction of MatrixNet in 2009.
With all new advancements that search engines make, machine learning doesn’t replace the variables and parameters that we’ve operated in before, but makes them better.
Like Google, Yandex has relied on a number of algorithms to improve search results for users, but since 2016 and the introduction of neural networks to their algorithm, Yandex has been building a much stronger algorithm.
How YATI Will Affect Yandex Optimization
Based on Yandex’s information and statements around the reveal of YATI at YaC2020, the new machine learning component of the algorithm will account for more than 50% of the final weighting.
This means that through better understanding of web documents and texts, making smaller changes to pages such as changing out title tags, adding in more keywords, and even exact match domains will no longer be as impactful (depending on competition and the niche).
As mentioned previously, this doesn’t mean that having strong technical, onpage, and offsite is now no longer needed. It now just makes it harder to game the system.
Can You Optimize For YATI?
As YATI is an evolution of Yandex’s algorithms, and not a revolution, so for the most part general Yandex optimization principles remain, and if anything have only been reinforced.
Fill Content Topic Gaps
Looking beyond keywords and at topics, you need to make sure that your content is as rich with them as your competitors.
A good example of this is if you’re trying to attract users looking to buy protein powders and meal replacement shakes, and you’re not talking about their ingredients, a breakdown of the calories, salts, and sugars, or how they’re manufactured but your competitors are, you’re an odd one out in the data set.
Structure Long Text Better
Breaking up pieces of text with headers can help users skim read and find relevant parts of the text they want to read, as well as add structure for search engines.
Based on the documentation around YATI, it’s widely thought within the Russian search community that breaking up text that is 250 to 300 words with a header can yield benefits.
What Is ERNIE (Baidu)?
Moving on from Yandex’s ML advancements, I’m going to look at ERNIE.
Baidu, like Google and Yandex, has a history with AI and machine learning, and in 2016 open-sourced the PaddlePaddle platform.
PaddlePaddle had been used internally by Baidu for a number of years to help develop algorithms and technologies to better their search product, scalable image classification, machine translation of texts, and their advertising platform.
ERNIE (version 1.0) was introduced into PaddlePaddle, and the wider Baidu ecosphere) in early 2019, with an updated version (2.0) coming around July that year.
XLNet being a joint venture between Google and Carnegie Mellon University, at and the time XLNet outperformed BERT.
As well as helping advance technology and search products, another great outcome of ERNIE is a system called DuTongChuan, which is the first ever context-aware simultaneous translation model.
ERNIE is an active part of the wider Baidu search algorithm, and is used to both general search results, and improve diversification within news feeds by removing duplicate stories (despite different headlines).
ERNIE also plays an active role in Baidu’s AI assistant, Xiao Du.
Using real time models, similar to DuTongChuan, Xiao Du uses ERNIE to better understand and more accurately respond to voice requests.
A lot of literature published around ERNIE is that of how it works, and processes data.
The actual impact it has had across Baidu search as a whole isn’t known, however, we also need to remember that Baidu SERP results populate in a very different way to both Google and Yandex do at this moment in time.
Baidu pulls through a number of rich snippets from it’s other products, such as Baike, Zhidao, and Tieba, meaning that organic queries may only be one or two results on the first page.
Can You Optimize For ERNIE?
Similar to other ML algorithms being deployed across search, ERNIE is an evolution of existing principles.
Baidu’s core algorithms (Money Plant, Pomegranate, Ice Bucket) have been encouraging webmasters to create better web experiences for users for a number of years, and now ERNIE is reinforcing these principles and rewarding websites who have invested in the user experience of search, and not just tried to game it.