Microsoft is tapping machine learning to curb ‘email overload’

Microsoft is tapping machine learning to curb ‘email overload’
Photo Credit: Photo Credit: Pixabay
11 Feb, 2019

Technology giant Microsoft is exploring ways of tapping machine learning to help employees of companies globally sift through the large amount of emails they receive on a daily basis, the company's research division said in a blog post.

According to a study conducted by McKinsey Global Institute, employees or professionals spend more than 28% of their time on email. And email is gaining more importance since it is not just a tool for communication but also acts as a repository of records and a tool for managing tasks, schedules, and collaborations.

“Identifying the emails you need to pay attention to is a challenging task,” said partner researcher and research manager Ryen White of Microsoft Research, who manages a team of about a dozen scientists and engineers and claims to receive 100-200 emails a day. “Right now, we end up doing a lot of that on our own.”

Microsoft's research division said that it has been studying the relationship between individuals and their emails for years and has been trying to find ways in which machine learning can better support users in their email responses and help make information in inboxes more accessible.

“We’re trying to bring in machine learning to make sense of a huge amount of data to make you more productive and efficient in your work,” said Microsoft senior researcher Ahmed Hassan Awadallah.

“Efficiency could come from a better ability to handle email, getting back to people faster, not missing things you would have missed otherwise,” he added.

Awadallah and his team have already interviewed 15 subjects and analysed the email logs of 40,000 anonymous users in a bid to understand the different reasons why a user would defer reading or replying to an email.

After compiling the results, Microsoft Research used the information to create datasets to train a machine learning model that could predict whether an email was deferred and for what reason.

According to Awadallah, the model has been successful in improving the email experience. One of the use-cases of the model is to remind users via email clients about deferred emails.

“If you have decided to leave an email for later, in many cases, you either just rely on memory or more primitive controls that your mail client provides like flagging your message or marking the message unread, and while these are useful strategies, we found that they do not provide enough support for users,” he explained.

White also said that they were working on improving smaller details of emails, especially the ones that contain collaborative requests, meetings, and follow-ups. While he said that existing tools such as Microsoft’s voice assistant Cortana were quite effective, there was still room for further advancement.