New spelling algorithm to reduce misspellings in Google Search

One in 10 queries on Google Search are misspelled and Google has now introduced a new spelling algorithm that uses a deep neural net to significantly improve the ability to decipher misspellings

One in 10 queries on Google Search are misspelled and the tech giant has now introduced a new spelling algorithm that uses a deep neural net to significantly improve the ability to decipher misspellings, in under three milliseconds.

 

 

According to Google, this single algorithm makes a greater improvement to spelling than all of its improvements over the last five years.

The company on Thursday detailed how artificial intelligence (AI) is powering its Search engine, announcing numerous improvements made to Search over the year and some new features coming soon.

Google has invested deeply in language understanding research and last year, it introduced how BERT language understanding systems are helping to deliver more relevant results in Google Search.

"We're excited to share that BERT is now used in almost every query in English, helping you get higher quality results for your questions," said Prabhakar Raghavan, Senior Vice President, Search and Assistant, Geo, Ads, Commerce, Payments & NBU.

Google recently made a breakthrough in ranking and is now able to not just index web pages but individual passages from the pages.

"This technology will improve 7 per cent of search queries across all languages as we roll it out globally," Raghavan added.

With new passage understanding capabilities, Google can understand that the specific passage is a lot more relevant to a specific query than a broader page on that topic.

As an example, if you search for "home exercise equipment," Google Search can now understand relevant subtopics, such as budget equipment, premium picks, or small space ideas, and show a wider range of content for you on the search results page.

"We'll start rolling this out by the end of this year," Raghavan said.

Using a new AI-driven approach, Google said it is now able to understand the deep semantics of a video and automatically identify key moments.

"This lets us tag those moments in the video, so you can navigate them like chapters in a book. Whether you're looking for that one step in a recipe tutorial, or the game-winning home run in a highlights reel, you can easily find those moments," the company informed.

Google thinks that by the end of this year, 10 per cent of searches will use this new technology.

--IANS

na/