Google’s spelling systems can help find the right word
Posted: Sun Dec 22, 2024 4:15 am
In the example below, Google can understand a query that is more similar to what a human would search for to show a more relevant result on a search for “Can you get medicine for someone pharmacy”.
Featured snippet example. Here’s an example from Google showing a featured snippet that’s most relevant to the query “Parking on a hill with no curb.” In the past, a search like this would have confused Google’s systems. Google said, “We overemphasized the word “curb” and ignored the word “no,” not realizing how critical that word was to adequately answering this query. So we’re returning results for parking on a hill with a curb.”
To all those who say or will say that RankBrain is whatsapp in philippines dead, Google says loudly no! RankBrain was Google's first AI method for understanding queries in 2015. It examines both queries and the content of web pages in Google's index to better understand what words mean.
BERT is not a replacement for RankBrain , it is an additional method for understanding content and queries. It is an addition to Google's ranking system . RankBrain can and will still be used for some queries. But when Google thinks that a query can be better understood with the help of BERT, Google will use it. In fact, a single query can use multiple methods, including BERT, to understand the query as best as possible.
Google explained that there are many ways to understand the meaning of a question and certainly each language can have one or more declensions.
For example, if you misspell something, to get what you need. And/or if you use a word that is a synonym for the actual word that is in the relevant documents, Google can detect that. BERT is another signal that Google uses to understand language. Depending on what you are searching for, any signal or combination of these signals could be used to understand your query and provide a relevant result.
Featured snippet example. Here’s an example from Google showing a featured snippet that’s most relevant to the query “Parking on a hill with no curb.” In the past, a search like this would have confused Google’s systems. Google said, “We overemphasized the word “curb” and ignored the word “no,” not realizing how critical that word was to adequately answering this query. So we’re returning results for parking on a hill with a curb.”
To all those who say or will say that RankBrain is whatsapp in philippines dead, Google says loudly no! RankBrain was Google's first AI method for understanding queries in 2015. It examines both queries and the content of web pages in Google's index to better understand what words mean.
BERT is not a replacement for RankBrain , it is an additional method for understanding content and queries. It is an addition to Google's ranking system . RankBrain can and will still be used for some queries. But when Google thinks that a query can be better understood with the help of BERT, Google will use it. In fact, a single query can use multiple methods, including BERT, to understand the query as best as possible.
Google explained that there are many ways to understand the meaning of a question and certainly each language can have one or more declensions.
For example, if you misspell something, to get what you need. And/or if you use a word that is a synonym for the actual word that is in the relevant documents, Google can detect that. BERT is another signal that Google uses to understand language. Depending on what you are searching for, any signal or combination of these signals could be used to understand your query and provide a relevant result.