Paper Reading #19: From Documents to Tasks: Deriving User Tasks from Document Usage Patterns

Commentary

See what I have to say about Shena's and Derek's work.

References

Brdiczka, O. (2010). From documents to tasks: deriving user tasks from document usage patterns. Proceeding of the Acm conference on intelligent user interfaces. . Hong Kong: http://www.iuiconf.org/.

Article Summary

In this paper, the researcher presents a novel approach to implementing a task management support system. A system such as this takes care of monitoring tasks of knowledge workers and clustering common tasks to increase productivity. The researcher argues that switching between multiple tasks is "expensive because each task requires some recovery time as well as the reconstitution of task context." This system is novel because it does not group documents based on title or content, both of which introduce privacy concerns. Instead, documents are applied a unique identifier and are filtered by their dwell time, or how long they have focus and are actively being accessed. Documents are then grouped by similarity via a spectral clustering algorithm. The proposed system had the additional benefit of not needing any user input whatsoever, a feature that is necessary for most other systems of its type. The system was evaluated over a period of a month, with observation days being non-contiguous. Normal knowledge workers were observed performing some commonly recurring tasks. The proposed system show a high level of effectiveness performing task grouping in comparison with similar systems.

Discussion

This paper seemed incredibly abstract to me. I'm sure that at least some of the industry-specific terms like "knowledge worker" and "recovery time" and "reconstitution of task context" must have meaning to someone out there, right? I just don't understand what the point of "task clustering" is supposed to be; what does it do? How does it increase productivity? To an lowly, uninitiated "luser" like me (NO I WILL NOT APOLOGIZE FOR THAT DON NORMAN), it feels like we're "promoting synergy" or some other dumb catch phrase:

Image courtesy of The Lonely Island and Oh! Ryan Kelley

I'm not knocking the paper, the author, or his work; he seems to have done a pretty stellar job. And I appreciate the novelty of his approach, not to mention the apparent success of his method. I just don't get it. At all.