Segmenting and Classifying Email Text

This site contains information and resources related to Andrew Lampert's email text segmentation and classification research.


This site contains information about Andrew Lampert's research into requests and commitments in workplace email messages. These messages form the crux of task-related email communication. In building machine-learning-based classifiers to automatically identify requests and commitments in email text, we have curated and developed a range of resources that may be useful for other researchers. We make many of these resources available for download from this site, including several annotated datasets that we have created. All datasets and code made available is licenced under a Creative Commons attribution licence for non-commercial use.

Currently, there are three main bodies of work featured on this website:

  1. Zebra, a system for automated email zoning;
  2. Request and Commitment Classification, primarily focused on an annotated dataset available for download; and
  3. A Microsoft Outlook Plug-in, a prototype integration of our request, commitment and zone classifiers with Microsoft Outlook 2007.

If you have any questions or comments, please contact Andrew Lampert.

Creative Commons License Resources made available on this site are licensed under a Creative Commons Attribution-Noncommercial 2.0 Generic License.