Discriminant Analysis: Classifying and Predicting Sexual Offenders
Bhajneet Kaur1, Laxmi Ahuja2, Vinay Kumar3
1Bhajneet Kaur*,AIIT, Amity University, Noida, (RDIAS), India.
2Dr. Laxmi Ahuja, AIIT, Amity University, Noida, India
3Dr. Vinay Kumar, Ex Scientist & Professor, Vice Chairman (Cum Chairman Elec) CSI Delhi Chapter, CSI Delhi Chapter, India.
Manuscript received on November 17., 2019. | Revised Manuscript received on November 24 2019. | Manuscript published on 30 November, 2019. | PP: 11939-11944 | Volume-8 Issue-4, November 2019. | Retrieval Number: D9851118419/2019©BEIESP | DOI: 10.35940/ijrte.D9851.118419
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Crimes or offences against women have their major contribution to spoil the physical and mental health of women. There are varieties of techniques and methods exists, through which crime against women can be analysed and predict for preventing and controlling it. In this paper a model has been proposed to classify and predict sexual offenders for two categories of victim namely major and minor using multiple discriminant analysis. IBM SPSS 21, statistical software has been used to build model with 85.3% of accuracy rate to classify and predict the cases. Dataset of sexual offenders has been extracted from the Chicago police portal system. After the compilation of the dataset four attributes have been used in the current analysis namely age, height, weight and victim_type. Three variables are taken as independent variables and one variable namely victim_type used as a target variable as it has two categories i.e. minor and major. All the three independent variables have their contribution in the model building but with different contribution rates. By using the standardized beta and wilk’s lambda, ranking of contribution of independent variables has been generated. Age of the sexual offenders has contributed the most to predict sexual offender. The second ranked variable found as weight, its contribution towards prediction is higher than “Height” but lower than “Age”. The variable “Height” got third rank as low predictor to predict sexual offenders for victim types. The proposed model is very effective and would be used to implement or taken the different security measures in the various public or private places to prevent and control the various kind of sexual offences.
Keywords: Sexual Offenders, Sexual Offences, Discriminant Analysis, Classification, Prediction, SPSS.
Scope of the Article: Classification.