Toward mixed-initiative email clustering

Organizing data into hierarchies is natural for humans. However, there is little work in machine learning that explores human-machine mixed-initiative approaches to organizing data into hierarchical clusters. In this pa-per we consider mixed-initiative clustering of a user’s email, in which the machine produces (initial and re-trained) hierarchical clusterings of email, and the user reviews and edits the initial hierarchical clustering, pro-viding constraints on the re-trained clustering model. Key challenges include (a) determining types of feed-back that users will find natural to provide, (b) develop-ing hierarchical clustering and retraining algorithms ca-pable of accepting these types of user feedback, (c) un-derstanding how machine’s clustering results and user feedback affect each other