Google will start serving more answers, as smartphone users don’t want blue links: VP Ben Gomes
Ben Gomes Google’s VP for Engineering and Search, explains how the depth of universal search and the context from Knowledge Graphs have helped to provide answers to people’s queries.
Thanks to Machine Learning (ML), Google search has been getting better and better.
If over the past year or so you have seen Googlethrowing up results like cricket scores and disease symptoms, that is because the search giant is clearly on a path to provide answers to people’s queries. “When you are on a cellphone, people actually want an answer to your query, not just blue links,” says Ben Gomes Google’s VP for Engineering and Search, explaining how the depth of universal search and the context from Knowledge Graphs were helping Mountain View do this.
Working on improving search, Google realised that a proper answer would have to be aided by concepts from the real world. “So we created Knowledge Graphs which contain a billion people, places and things and about 70 billion connections between them. For instance, if you ask who is the prime minister of India, we know there is a current prime minister, what his name is, we know how he is connected to other people and can thus answer your question in a natural way,” Gomes explains toindianexpress.com on the sidelines of the Google for India event.
Gomes, who has been with Google for 19 years, before “even people in California knew the name”, says Knowledge Graph was created to answer questions the way people are expecting them to be answered and also give answers to all the different kinds of things they might ask. “That allows us to go from the ten blue links on the desktop to what they were beginning to expect on a mobile phone which very different, much richer experience.”
Over the years, Google Search has evolved tremendously. Now, it can give you answers even before you ask via the feed. “There are a lot of cases were we can give answers even before the question has been asked, like the time to the airport, or if your flight is delayed. And there are other cases where we need to know what is needed. And needs vary dramatically. With a feed we can give you results on things of interest to you over time.”
A lot of this is also because of technologies like Machine Learning (ML) becoming better. Gomes says that many of the hundreds of signals in search are now becoming machine learnt. “Gradually, we would be using machine learning in many different parts of search from language understanding to how we combine the various signals. There are different types of signals; pagerank being one, then there are words on a page, fonts, who points to this page and what is their page rank and so on. There are many different signals that go into evaluating if this page is a good result for this query and better than another page. Each of these signals can have a machine learning component to it,” he explains.
How Google tackled fake news
Also Read: LG G7 might come with advanced iris scanner for facial recognition, patent reveals.
The past year has been a challenging one for Google Search and Facebook because of the controversy around fake news in the US. Gomes says it is actually a small problem as only about 2 per cent of the queries are affected, “but these are important queries and we are very bothered by it”. Google reacted by changing the rater guidelines. “Whenever we make a change in the search algorithm, we show the 10,000-odd raters around A and B and ask them which is better. We ask them to judge based on rater guidelines, which is essentially a description of what search does. Using these raters, we change the algorithm, hopefully making them better and better.
Also Check for Update Button on Pixel, Nexus Devices to Finally Work in 2018.
Gomes accepts that they were under the impression that news would come from good sources. “Now, for those kind of queries there are two things we look at: how relevant is the result and how authoritative is the source. The raters look at both and now we have asked them to weigh the authoritativeness of the source more than how exactly the words match,” he says, adding that Google launched something like 2000 changes in the last year resulting in a huge improvement in the kind of queries people were seeing a problem with last year.
But Gomes knows there is no way to make this problem go away completely. “Because every day we see 15 per cent of our queries which we have never seen before; there are millions of new documents. There are millions of things that will happen… people who will deceive the algorithm.” However, that is nothing new for Google, as people have been trying to game the system right from the early days of page ranking. “It is not new in that sense, we are just taking new approaches to tackling the particular problem.
To get our top Technology News, Gadget review, Latest phone stories delivered to your inbox, kindly sign up for the News letter programme.
Thanks to Machine Learning (ML), Google search has been getting better and better.
Working on improving search, Google realised that a proper answer would have to be aided by concepts from the real world. “So we created Knowledge Graphs which contain a billion people, places and things and about 70 billion connections between them. For instance, if you ask who is the prime minister of India, we know there is a current prime minister, what his name is, we know how he is connected to other people and can thus answer your question in a natural way,” Gomes explains toindianexpress.com on the sidelines of the Google for India event.
Gomes, who has been with Google for 19 years, before “even people in California knew the name”, says Knowledge Graph was created to answer questions the way people are expecting them to be answered and also give answers to all the different kinds of things they might ask. “That allows us to go from the ten blue links on the desktop to what they were beginning to expect on a mobile phone which very different, much richer experience.”
Over the years, Google Search has evolved tremendously. Now, it can give you answers even before you ask via the feed. “There are a lot of cases were we can give answers even before the question has been asked, like the time to the airport, or if your flight is delayed. And there are other cases where we need to know what is needed. And needs vary dramatically. With a feed we can give you results on things of interest to you over time.”
A lot of this is also because of technologies like Machine Learning (ML) becoming better. Gomes says that many of the hundreds of signals in search are now becoming machine learnt. “Gradually, we would be using machine learning in many different parts of search from language understanding to how we combine the various signals. There are different types of signals; pagerank being one, then there are words on a page, fonts, who points to this page and what is their page rank and so on. There are many different signals that go into evaluating if this page is a good result for this query and better than another page. Each of these signals can have a machine learning component to it,” he explains.
How Google tackled fake news
Also Read: LG G7 might come with advanced iris scanner for facial recognition, patent reveals.
The past year has been a challenging one for Google Search and Facebook because of the controversy around fake news in the US. Gomes says it is actually a small problem as only about 2 per cent of the queries are affected, “but these are important queries and we are very bothered by it”. Google reacted by changing the rater guidelines. “Whenever we make a change in the search algorithm, we show the 10,000-odd raters around A and B and ask them which is better. We ask them to judge based on rater guidelines, which is essentially a description of what search does. Using these raters, we change the algorithm, hopefully making them better and better.
Also Check for Update Button on Pixel, Nexus Devices to Finally Work in 2018.
Gomes accepts that they were under the impression that news would come from good sources. “Now, for those kind of queries there are two things we look at: how relevant is the result and how authoritative is the source. The raters look at both and now we have asked them to weigh the authoritativeness of the source more than how exactly the words match,” he says, adding that Google launched something like 2000 changes in the last year resulting in a huge improvement in the kind of queries people were seeing a problem with last year.
But Gomes knows there is no way to make this problem go away completely. “Because every day we see 15 per cent of our queries which we have never seen before; there are millions of new documents. There are millions of things that will happen… people who will deceive the algorithm.” However, that is nothing new for Google, as people have been trying to game the system right from the early days of page ranking. “It is not new in that sense, we are just taking new approaches to tackling the particular problem.
To get our top Technology News, Gadget review, Latest phone stories delivered to your inbox, kindly sign up for the News letter programme.
Comments
Post a Comment