• DocumentCode
    495486
  • Title

    How Emergent Self Organizing Maps Can Help Counter Domestic Violence

  • Author

    Poelmans, Jonas ; Elzinga, Paul ; Viaene, Stijn ; Van Hulle, Marc M. ; Dedene, Guido

  • Author_Institution
    Fac. of Bus. & Econ., K.U. Leuven, Leuven, Belgium
  • Volume
    4
  • fYear
    2009
  • fDate
    March 31 2009-April 2 2009
  • Firstpage
    126
  • Lastpage
    136
  • Abstract
    Topographic maps are an appealing exploratory instrument for discovering new knowledge from databases. During the past years, new types of self organizing maps (SOM) were introduced in the literature, including the recent emergent SOM. The ESOM is used to study a large set of police reports describing a whole range of violent incidents that occurred during the year 2007 in the police region Amsterdam-Amstelland (the Netherlands). It is demonstrated that it provides an exploratory search instrument for examining unstructured text in police reports. First, it is shown how the ESOM was used to discover a whole range of new features that better distinguish domestic from non-domestic violence cases. Then, it is demonstrated how this resulted in a significant improvement in classification accuracy. Finally, the ESOM is showcased as a powerful instrument for the domain expert interested in an in-depth investigation of the nature and scope of domestic violence.
  • Keywords
    cartography; data mining; police data processing; self-organising feature maps; classification accuracy; databases; domestic violence; emergent self organizing map; exploratory search instrument; knowledge discovery; police region Amsterdam-Amstelland; police report; topographic map; Computer science; Counting circuits; Data engineering; Databases; Engineering management; Instruments; Knowledge engineering; Knowledge management; Power generation economics; Self organizing feature maps; Emergent Self Organizing Maps; domestic violence; exploratory data analysis; knowledge discovery;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Information Engineering, 2009 WRI World Congress on
  • Conference_Location
    Los Angeles, CA
  • Print_ISBN
    978-0-7695-3507-4
  • Type

    conf

  • DOI
    10.1109/CSIE.2009.299
  • Filename
    5170974