Intelligent Collaboration Tools Lab

← Back to laboratory

Recommending Messages in a Software Engineering Messenger Workspace

The Smart Message Feed Bot provides a personalized digest of messages in public Space channels. It is meant to help Space users navigate through all of the messages without missing vital information.

Developers today spend at least 20% of their time communicating in various forms. Much of this communication happens through messengers. The issue is that in larger organizations it is beyond the scope of human attention to follow all potentially relevant messages.

We are building a message recommendation system to aid professionals in keeping up with the information flow across the messenger workspace. To our knowledge, this is the first attempt to create an automated message recommendation tool for channel-based business messengers.

We implemented a recommendation model and deployed it as a bot application in a JetBrains internal workspace. The initial feedback we received shows that the concept of message recommendation is a promising direction for making collaboration tools more intelligent.

Currently, we are working on improving the recommendation algorithm and integrating the bot deeper into Space.

Participants

Vladimir Kovalenko
Ekaterina Koshchenko

Publications

May 2022
Ekaterina Koshchenko, Egor Klimov, Vladimir Kovalenko

Multimodal Recommendation of Messenger Channels

Collaboration platforms, such as GitHub and Slack, are a vital instrument in the day-to-day routine of software engineering teams. The data stored in these platforms has a significant value for datadriven methods that assist with decision-making and help improve software quality.