A Decision Tree Approach to Predicting Recidivism in Domestic Violence

  • Senuri Wijenayake
  • Timothy Graham
  • Peter ChristenEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11154)


Domestic violence (DV) is a global social and public health issue that is highly gendered. Being able to accurately predict DV recidivism, i.e., re-offending of a previously convicted offender, can speed up and improve risk assessment procedures for police and front-line agencies, better protect victims of DV, and potentially prevent future re-occurrences of DV. Previous work in DV recidivism has employed different classification techniques, including decision tree (DT) induction and logistic regression, where the main focus was on achieving high prediction accuracy. As a result, even the diagrams of trained DTs were often too difficult to interpret due to their size and complexity, making decision-making challenging. Given there is often a trade-off between model accuracy and interpretability, in this work our aim is to employ DT induction to obtain both interpretable trees as well as high prediction accuracy. Specifically, we implement and evaluate different approaches to deal with class imbalance as well as feature selection. Compared to previous work in DV recidivism prediction that employed logistic regression, our approach can achieve comparable area under the ROC curve results by using only 3 of 11 available features and generating understandable decision trees that contain only 4 leaf nodes.


Crime prediction Re-offending Class imbalance Feature selection 


  1. 1.
    Australian and New Zealand Society of Criminology.
  2. 2.
    Australian Bureau of Statistics: Personal safety survey 2016 (2017)Google Scholar
  3. 3.
    Australian Institute of Health and Welfare: Family, domestic and sexual violence in Australia (2018)Google Scholar
  4. 4.
    Australian Institute of Health and Welfare; Australia’s National Research Organisation for Women’s Safety Limited (ANROWS): Examination of the health outcomes of intimate partner violence against women: state of knowledge paper (2016)Google Scholar
  5. 5.
    Boxall, H., Rosevear, L., Payne, J.: Identifying first time family violence perpetrators: the usefulness and utility of categorisations based on police offence records. Trends Issues Crime Crim. Justice (487) (2015).;dn=370297168229940;res=IELHSS
  6. 6.
    Bulmer, C.: Australian police deal with a domestic violence matter every two minutes. ABC News (2015)Google Scholar
  7. 7.
    Cox, P.: Violence against women in Australia: additional analysis of the Australian Bureau of Statistics’ personal safety survey. Horizons Research Report, Australia’s National Research Organisation for Women’s Safety (ANROWS) (2012)Google Scholar
  8. 8.
    Drummond, C., Holte, R.C.: C4.5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling. In: ICML Workshop (2003)Google Scholar
  9. 9.
    Fawcett, T.: ROC graphs: notes and practical considerations for researchers. Technical Report HPL-2003-4, HP Laboratories, Palo Alto (2004)Google Scholar
  10. 10.
    Fitzgerald, R., Graham, T.: Assessing the risk of domestic violence recidivism. Crime and Justice Bulletin, NSW Bureau of Crime Statistics and Research (2016)Google Scholar
  11. 11.
    Gillingham, P.: Risk assessment in child protection: problem rather than solution? Aust. Soc. Work 59(1), 86–98 (2006)CrossRefGoogle Scholar
  12. 12.
    Hand, D.: Classifier technology and the illusion of progress. Stat. Sci. 21(1), 1–14 (2006)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Hsieh, T., Wang, Y.H., Hsieh, Y.S., Ke, J.T., Liu, C.K., Chen, S.C.: Measuring the unmeasurable: a study of domestic violence risk prediction and management. J. Technol. Hum. Serv. 36, 56–68 (2018)CrossRefGoogle Scholar
  14. 14.
    Japkowicz, N., Stephen, S.: The class imbalance problem: a systematic study. Intell. Data Anal. 6(5), 429–449 (2002)CrossRefGoogle Scholar
  15. 15.
    Krug, E., Mercy, J., Dahlberg, L., Zwi, A.: The world report on violence and health (2002)Google Scholar
  16. 16.
    Lior, R., Maimon, O.: Data Mining with Decision Trees: Theory and Applications, vol. 81, 2nd edn. World Scientific, Singapore (2014)zbMATHGoogle Scholar
  17. 17.
    Mason, R., Julian, R.: Analysis of the Tasmania police risk assessment screening tool (RAST). Final report, Tasmanian Institute of Law Enforcement Studies, University of Tasmania (2009)Google Scholar
  18. 18.
    Messing, J.T., Campbell, J., Sullivan Wilson, J., Brown, S., Patchell, B.: The lethality screen: the predictive validity of an intimate partner violence risk assessment for use by first responders. J. Interpers. Violence 32(2), 205–226 (2017)CrossRefGoogle Scholar
  19. 19.
    Neuilly, M.A., Zgoba, K.M., Tita, G.E., Lee, S.S.: Predicting recidivism in homicide offenders using classification tree analysis. Homicide Stud. 15(2), 154–176 (2011)CrossRefGoogle Scholar
  20. 20.
    New South Wales Police Force: NSW Police Force Annual Report, 2014–15 (2015)Google Scholar
  21. 21.
    NSW Bureau of Crime Statistics and Research: Re-offending statistics for NSW (2018)Google Scholar
  22. 22.
    Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., et al.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12(Oct), 2825–2830 (2011)MathSciNetzbMATHGoogle Scholar
  23. 23.
    Quinlan, J.R.: Induction of decision trees. Mach. Learn. 1(1), 81–106 (1986)Google Scholar
  24. 24.
    Rice, M.E., Harris, G.T., Hilton, N.: The violence risk appraisal guide and sex offender risk appraisal guide for violence risk assessment. In: Handbook of Violence Risk Assessment, pp. 99–119. Routledge, Abington (2010)Google Scholar
  25. 25.
    Ringland, C., et al.: Measuring recidivism: police versus court data. BOCSAR NSW Crime Justice Bull. (175), 12 (2013).;dn=344533455169983;res=IELHSS
  26. 26.
    Ronda, M.: Recividism. In: Encyclopedia of Social Problems, pp. 756–757. Sage Publications, Inc. (2008)Google Scholar
  27. 27.
    Wall, L.: Gender equality and violence against women: what’s the connection? Australian Institute of Family Studies, ACSSA Research Summary, no. 7 (2014)Google Scholar
  28. 28.
    Western Australia Police Force: Crime in Western Australia (2018)Google Scholar
  29. 29.
    Wu, L., Haynes, M., Smith, A., Chen, T., Li, X.: Generating life course trajectory sequences with recurrent neural networks and application to early detection of social disadvantage. In: Cong, G., Peng, W.-C., Zhang, W.E., Li, C., Sun, A. (eds.) ADMA 2017. LNCS (LNAI), vol. 10604, pp. 225–242. Springer, Cham (2017). Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Senuri Wijenayake
    • 1
  • Timothy Graham
    • 2
  • Peter Christen
    • 2
    Email author
  1. 1.Faculty of Information TechnologyUniversity of MoratuwaMoratuwaSri Lanka
  2. 2.Research School of Computer ScienceThe Australian National UniversityCanberraAustralia

Personalised recommendations