Police Forensic science performance indicators – a new approach to data validation

  • R. Adderley
  • J W Bond

Abstract

DNA and fingerprint identifications continue to form an integral part of the detection of a wide range of crime types, especially volume crime such as burglary and auto crime. More than ten years ago, researchers first commented on the lack of emphasis on ‘outcome’ (i.e. crime detection) related performance indicators for UK police forces. Since then much work has been carried out, mainly by the Association of Chief Police Officers of England & Wales and the Home Office, to produce a framework of forensic science performance indicators that reflect accurately the contribution made by forensic science to crime detection. In this paper, we consider the data currently being collected by five UK police forces that use popular proprietary computer based data collection systems. The accuracy of the data collection has been analysed using a neural network and has identified collection errors in all five forces. These errors are such that they could adversely affect the accuracy and interpretation of the national collection of forensic science data conducted by the Home Office. We propose using this neural network to check the accuracy of data collection and also to provide a ‘front end’ collator for national forensic science data returns to the Home Office. Such an approach would improve the accuracy of data collection nationally and also provide some reassurance over the consistency of data recording by individual forces.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Touche Ross. Review of scientific support for the police, 3 vols, London: Home Office, 1987.Google Scholar
  2. [2]
    Police Requirements Support Unit, Scientific Support Team. Recommendations for the performance monitoring of scientific support to the police service, London: Home Office, 1991.Google Scholar
  3. [3]
    Tilley N, Ford A. Forensic science and criminal investigation. Crime Detection and Prevention Paper 73, London: Home Office, 1996.Google Scholar
  4. [4]
    McCulloch H. Police use of forensic science. Police Research Series Paper 19, London: Home Office, 1996.Google Scholar
  5. [5]
    Association of Chief Police Officers & Forensic Science Service. Using Forensic Science Effectively. London: ACPO, 1996.Google Scholar
  6. [6]
    Sims, C. Output related performance indicators for scientific support. Private communication, 2001.Google Scholar
  7. [7]
    Her Majesty’s Inspectorate of Constabulary. Under the Microscope. London: ACPO, 2000.Google Scholar
  8. [8]
    Her Majesty’s Inspectorate of Constabulary. Under the Microscope Refocused. London: ACPO, 2002.Google Scholar
  9. [9]
    Williams, S. R. The Management Of Crime Scene Examination In Relation To The Investigation Of Burglary And Vehicle Crime. Home Office online report 24/04. London: Home Office, 2004.Google Scholar
  10. [10]
    Police Standards Unit. Forensic Performance Monitors, London: Home Office, 2006.Google Scholar
  11. [11]
    Adriaans, P., & Zantinge, D. Data Mining, New York, Addison-Wesley, 1996.Google Scholar
  12. [12]
    Chapman, P., Clinton, J., Kerber, R., et al. CRISP-DM 1.0 Step-by-step data mining guide, USA: SPSS Inc. CRISPWP-0800, 2000.Google Scholar
  13. [13]
    Giraud-Carrier, C., Povel, O. Characterising data mining software. Journal of Intelligent Data Analysis 2003; 7(3) pp. 181-192.Google Scholar
  14. [14]
    Mena, J. Investigative data mining for security and criminal detection, 2003.Google Scholar
  15. [15]
    Adderley, R., Bond, J.W., Townsley, M. Use of data mining techniques to model crime scene investigator performance 26th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, Cambridge UK, 2006.Google Scholar
  16. [16]
    Adderley, R., Bond, J.W. The effects of deprivation on the time spent examining crime scenes and the recovery of DNA & fingerprints. J. Forensic Sci. 2007 (in print).Google Scholar
  17. [17]
    Adderley, R., Bond, J.W., Townsley, M. Predicting crime scene attendance. Int. J. Police Science & Management, 2007 (in print).Google Scholar
  18. [18]
    Home Office. National Policing Plan 2005-2008, London: Home Office, 2004.Google Scholar
  19. [19]
    Swingler, K. Applying Neural networks; A practical guide. Morgan Kaufman, San Francisco, 1996.Google Scholar
  20. [20]
    Duda, R.O., Hart, P.E. Pattern analysis and scene analysis. John Wiley, New York, 1971.Google Scholar
  21. [21]
    Langley, P., Sage, S. Induction of selective Bayesian classifiers, Proceedings 10th Conference on Uncertainty in Artificial Intelligence, Seattle, WA; Morgan Kaufmann, 1994, pp. 339-406.Google Scholar

Copyright information

© Springer-Verlag London Limited 2008

Authors and Affiliations

  • R. Adderley
    • 1
  • J W Bond
    • 2
  1. 1.A E Solutions (BI)>
  2. 2.Northamptonshire PoliceUK

Personalised recommendations