The new Google BERT update will help to improve 1 out of 10 search queries by better contextualizing word meaning through AI…
Google says it’s just introduced an algorithmic update based on artificial intelligence which incorporates natural language processing. If that sounds complex, it is. But then again, so is human communication.
It’s called BERT, which stands for Bidirectional Encoder Representations from Transformers.
Google BERT Update Injects Natural Language Processing System to Improve 10 Percent of Search Queries
Google states this is a milestone in better parsing search queries. The search engine will now be able to better consider the full context of sentences. It’s doing this by more correctly assessing words which come before and after keywords.
Here’s an example. For the query “2019 brazil traveler to usa need a visa” Google would essentially throw away the preposition “to.” So, it would return listings which didn’t match the intent of the search.
Now, the search engine will no longer ignore certain words and attempt to understand the entire query. The old Google search algorithm treated that sentence as a “bag of words,” according to Pandu Nayak, Google fellow and VP of search.
With the update, the search engine won’t discard words but try to understand and contextualize the entire query.
Nayak writes the following on the official Google blog:
“Last year, we introduced and open-sourced a neural network-based technique for natural language processing (NLP) pre-training called Bidirectional Encoder Representations from Transformers, or as we call it–BERT, for short. This technology enables anyone to train their own state-of-the-art question answering system.
This breakthrough was the result of Google research on transformers: models that process words in relation to all the other words in a sentence, rather than one-by-one in order. BERT models can therefore consider the full context of a word by looking at the words that come before and after it—particularly useful for understanding the intent behind search queries.”