Applied Intelligence

, Volume 36, Issue 2, pp 308–319

Case learning for CBR-based collision avoidance systems

  • Yuhong Liu
  • Chunsheng Yang
  • Yubin Yang
  • Fuhua Lin
  • Xuanmin Du
  • Takayuki Ito
Article

DOI: 10.1007/s10489-010-0262-z

Cite this article as:
Liu, Y., Yang, C., Yang, Y. et al. Appl Intell (2012) 36: 308. doi:10.1007/s10489-010-0262-z

Abstract

With the rapid development of case-based reasoning (CBR) techniques, CBR has been widely applied to real-world applications such as collision avoidance systems. A successful CBR-based system relies on a high-quality case base, and a case creation technique for generating such a case base is highly required. In this paper, we propose an automated case learning method for CBR-based collision avoidance systems. Building on techniques from CBR and natural language processing, we developed a methodology for learning cases from maritime affair records. After giving an overview on the developed systems, we present the methodology and the experiments conducted in case creation and case evaluation. The experimental results demonstrated the usefulness and applicability of the case learning approach for generating cases from the historic maritime affair records.

Keywords

Case-based reasoning Ship collision avoidance Maritime affair records Case learning Case base management 

Copyright information

© Springer Science+Business Media, LLC 2010

Authors and Affiliations

  • Yuhong Liu
    • 1
  • Chunsheng Yang
    • 2
  • Yubin Yang
    • 3
  • Fuhua Lin
    • 4
  • Xuanmin Du
    • 5
  • Takayuki Ito
    • 6
  1. 1.Shanghai Aerospace Electronic Technology InstituteShanghaiChina
  2. 2.Institute for Information TechnologyNational Research Council CanadaOttawaCanada
  3. 3.State Key Laboratory for Novel Software TechnologyNanjing UniversityNanjingChina
  4. 4.Athabasca UniversityAthabascaCanada
  5. 5.Shanghai Marine Electronic Equipment Research InstituteShanghaiChina
  6. 6.Nagoya Institute of TechnologyNagoyaJapan