How Google using machine learning to understand Gmail users

Google AI is constantly evolving its pattern to learn more about its customer.

Last year they introduced Federated Learning on Gboard. When Gboard shows a suggested query, your phone locally stores information about the current context and whether you clicked the suggestion. Federated Learning processes that history on-device to suggest improvements to the next iteration of Gboard’s query suggestion model.

It actually enables a shared learning between phone and cloud. First your phone downloads the current model, improves it by learning from data on your phone, and then summarizes the changes as a small focused update. Only this update to the model is sent to the cloud, using encrypted communication, where it is immediately averaged with other user updates to improve the shared model. All the training data remains on your device, and no individual updates are stored in the cloud. 

Also read : How to get rank in google locally?

But how Google using machine learning in Gmail

Did you see the recent update of Gmail!

I think you did, right ?

Meh! what do we care?

I was given the same answer from past couple of month. But the truth is Google machine learning is actually understanding your Gmail context in Gmail apps.

When somebody emails (simple context mail) you, you would experience three smart reply suggestions as per your Mail subject, Mail Body and Mail attachment.

Let me give you some examples:

Example 1:

Now I am explaining this image.

The red boxed one is mail subject. As per the mail subject “Test“, it understands this is a test mail and Gmail suggests “It works!” as smart reply.

Then the dark blue boxed one is mail body, which is written “How are you“. In reply to that particular segment Gmail suggests “I’m good, how about you?” as smart reply.

And the third one is green boxed which is written “That was pretty hilarious” and the Gmail smart reply suggests me “Haha, thanks“.

Example 2:

Example 3:

But this suggestion won’t appear in gibberish mail or email marketing template.

Google gradually extends this feature from Gmail to Allo and Android Messages. Google is also using Neural Learning AI in Google Translator and Deep Learning AI to detect breast cancer. To know more about Google AI, click this link.