Accession Number:

ADA585927

Title:

Accelerating Human-Computer Collaborative Search through Learning Comparative and Predictive User Models

Descriptive Note:

Conference paper

Corporate Author:

CALIFORNIA UNIV SANTA CRUZ

Personal Author(s):

Report Date:

2012-07-09

Pagination or Media Count:

10.0

Abstract:

Interactive Evolutionary Algorithms IEAs are one of the few systems in which a human user and a computer algorithm are collaboratively working on a problem. To turn a basic IEA into the start of a Human-Computer Collaborative Computational system we have developed a system called The Approximate User TAU. With TAU, as the user interacts with the IEA a model of the users preferences is constructed and continually refined and it is this user-model which drives search. Here two variations of a user-modeling approach are compared to determine if this approach can accelerate IEA search. The two user-modeling approaches compared are 1 learning a classifier which correctly determines which of two designs is better and 2 learning a model which predicts a fitness score. Rather than having people do the user-testing, we propose the use of a simulated user as an easier means to test IEAs. Both variants of the TAU IEA are compared against a basic IEA and it is shown that TAU is up to 2.7 times faster and 15 times more reliable at producing near optimal results.

Subject Categories:

  • Administration and Management
  • Psychology
  • Biology
  • Computer Programming and Software
  • Mechanics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE