Mixer: mixed-initiative data retrieval and integration by example

  • Authors:
  • Steven Gardiner;Anthony Tomasic;John Zimmerman;Rafae Aziz;Kathryn Rivard

  • Affiliations:
  • Carnegie Mellon School of Computer Science, Pittsburgh PA;Carnegie Mellon School of Computer Science, Pittsburgh PA;Carnegie Mellon School of Computer Science, Pittsburgh PA;Carnegie Mellon School of Computer Science, Pittsburgh PA;Carnegie Mellon School of Computer Science, Pittsburgh PA

  • Venue:
  • INTERACT'11 Proceedings of the 13th IFIP TC 13 international conference on Human-computer interaction - Volume Part I
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Office administrators are frequently asked to create ad hoc reports based on web accessible data. The web contains the desired data but does not allow efficient access in the way the administrator needs, prompting a tedious and labor-intensive task of retrieving and integrating the required data. Mixer is a programming-by-demonstration (PBD) tool empowering administrators to construct ad hoc reports from diverse web sources without tedious piecemeal labor. Mixer's design builds on the exploration into end user conceptualization of data retrieval tasks from our previous Wizard-of-Oz study [39], and incorporates insights from mixed-initiative researchers into collaboration between end users and software agents. This paper justifies the design decisions that drive Mixer, focusing on general lessons for designers of programming-bydemonstration systems targeting nonprogrammers. We evaluate Mixer by performing a user study showing that administrators are able to accomplish programming tasks without needing to understand programming concepts for data retrieval and integration.