Optimizing Question-Answering Systems Using Genetic Algorithms

PDF

Authors
  1. Allard, U.C.
  2. Khoury, R.
  3. Lamontage, L.
  4. Bergeron, J.
  5. Laviolette, F.
  6. Bergeron-Guyard, A.
Corporate Authors
Defence Research and Development Canada, Valcartier Research Centre, Quebec QC (CAN)
Abstract
In this paper, we consider the challenge of optimizing the behaviour of a question-answering system that can adapt its sequence of processing steps to meet the information needs of a user. One problem is that the sheer number of possible processing sequences the system could use makes it impos- sible to conduct a complete search for the optimal sequence. Instead, we have developed a genetic algorithm to explore the space of possible sequences. Our results show that this approach gives the system the adaptability we desire while still performing better than a human-optimized system.
Keywords
genetic algorithms;question answering;intelligent virtual assistant
Report Number
DRDC-RDDC-2015-P003 — Scientific Letter
Date of publication
21 Apr 2015
Number of Pages
6
DSTKIM No
CA040320
CANDIS No
801295
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: