Smart To-Do: Automatic Generation of To-Do Items from Emails

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8680–8689 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 8680 Smart To-Do: Automatic Generation of To-Do Items from Emails Sudipto Mukherjee†∗ Subhabrata Mukherjee‡ Marcello Hasegawa‡ Ahmed Hassan Awadallah‡ Ryen White‡ †University of Washington, Seattle sudipm@uw.edu,‡Microsoft Research AI {submukhe, marcellh, hassanam, ryenw}@microsoft.com Abstract Intelligent features in email service applica- tions aim to increase productivity by help- ing people organize their folders, compose their emails and respond to pending tasks. In this work, we explore a new application, Smart-To-Do, that helps users with task man- agement over emails. We introduce a new task and dataset for automatically generating To-Do items from emails where the sender has promised to perform an action. We de- sign a two-stage process leveraging recent ad- vances in neural text generation and sequence- to-sequence learning, obtaining BLEU and ROUGE scores of 0.23 and 0.63 for this task. To the best of our knowledge, this is the first work to address the problem of composing To- Do items from emails. 1 Introduction Email is one of the most used forms of communi- cation especially in enterprise and work settings (Radicati and Levenstein, 2015). With the grow- ing number of users in email platforms, service providers are constantly seeking to improve user experience for a myriad of applications such as online retail, instant messaging and event manage- ment (Feddern-Bekcan, 2008). Smart Reply (Kan- nan et al., 2016) and Smart Compose (Chen et al., 2019) are two recent features that provide contex- tual assistance to users aiming to reduce typing efforts. Another line of work in this direction is for automated task management and scheduling. For example. the recent Nudge feature1 in Gmail and Insights in Outlook2 are designed to remind users to follow-up on an email or pay attention to pending tasks. Smart To-Do takes a step further in task assis- tance and seeks to boost user productivity by auto- matically generating To-Do items from their email ∗Work done as an intern at Microsoft Research. 1 Gmail Nudge 2 Outlook Insights From: Alice To: john@contoso.com Subject: Sales Report Hi John, From: John To: alice@contoso.com Subject: RE: Sales Report I am doing well. Thanks! I am travelling now. I will send it to you once I am back. Send the product launch sales report to Alice How are you? I wanted to follow up on our previous product launch meeting. Could you send me the sales report you mentioned? I want to forward it to my manager and others in the team. Best, Alice Hi Alice, -John Figure 1: An illustration showing the email and a com- mitment sentence (in yellow) and the target To-Do item, along with other email meta-data. context. Text generation from emails, like creating To-Do items, is replete with complexities due to the diversity of conversations in email threads, hetero- geneous structure of emails and various meta-deta involved. As opposed to prior works in text gener- ation like news headlines, email subject …

Content truncated to 500 words