Intelligent Contextual Task Support for Mails

Abstract

Due to its long existence and its wide spread, in most companies email is amongst the usual communication mediums and therefore commonly used for cross-enterprise communication. This is also the case in the scientific context, where projects are often run by big, international teams and have stakeholders with different roles and points of view. Therefore, communication needs are even higher leading to many emails being exchanged back and forth.

This causes multiple problems. For a start, all emails that are received arrive in one central inbox, making it difficult to keep track of individual projects and conversations. Furthermore many mails contain unknown terms like foreign words and concepts outside of the receiver’s subject area or mention the names of unknown persons. Besides that, keeping track of tasks that are assigned via email is hard to do, especially when the tasks are relevant to multiple people, have explicit due dates or consist of multiple subtasks.

Some existing mail clients solve parts of these problems by e.g. providing an automated labeling system that takes care of sorting the incoming mails. For other problems like collaborative task management or foreign word lookup there are commercial tools available but these are usually designed to be standalone solutions and do not integrate well with email clients. A possible solution to the overall problem could be the Intelligent Contextual Mail (ICM) Client that is developed at the SEBIS chair and currently exists as a prototype. It tries to assist users by analyzing incoming emails and suggesting contextual information in a dedicated sidebar. It also integrates multiple services like Wikipedia for looking up unknown concepts or Trello for creating and managing tasks. This removes the need to switch between different tools as everything can be done from within the mail client.

The goal of this Guided Research is to compare existing email applications regarding their contextual task support. The identified strengths and weaknesses of the individual applications can then be contrasted with the concepts of ICM and with the current prototypical application. This enables the creation of a prioritized feature roadmap for future development efforts. The second goal of this work is the extension of the ICM prototype to properly support more than one task provider by introducing a new conceptual layer for task management. The developed concept will then be proofed to work by integrating SocioCortex, a model-based content management platform, as an optional task provider.

Keywords: ICM, Connecare, Tasks, SocioCortex

Core literature

• Andrew Faulring, Brad Myers, Ken Mohnkern, Bradley Schmerl, Aaron Steinfeld, John Zimmerman, Asim Smailagic, Jeffery Hansen, and Daniel Siewiorek. 2010. Agent-assisted task management that reduces email overload. In Proceedings of the 15th international conference on Intelligent user interfaces - IUI '10. ACM Press.
• Andrew Lampert, Robert Dale, and Cecile Paris. 2010. Detecting emails containing requests for action. In Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics (HLT ’10). Association for Computational Linguistics, Stroudsburg, PA, USA, 984–992.
• Nicholas Kushmerick and Tessa Lau. 2005. Automated email activity management. In Proceedings of the 10th international conference on Intelligent user interfaces - IUI '05. ACM Press.
• Colin Shearer. 2000. The CRISP-DM model: the new blueprint for data mining. Journal of Data Warehousing 5 (01 2000), 13–22.