The CR–Ω+ Classification Algorithm for Spatio–Temporal Prediction of Criminal Activity

 

S. Godoy–Calderón², H. Calvo^*1,2, V. M. Martínez–Hernández², M. A. Moreno–Armendáriz²

 

¹ Nara Institute of Science and Technology, Takayama, Ikoma, Nara 630–0192, Japan sgodoyc@cic.ipn.mx, *hcalvo@cic.ipn.mx, *calvo@is.naist.jp mhernandezb07@sagitario.cic.ipn.mx, marco_moreno@cic.ipn.mx.

² Centro de Investigación en Computación CIC–IPN. Av. Juan de Dios Bátiz s/n esq. Manuel Othón de Mendizábal, 07738, Mexico City, Mexico.

 

ABSTRACT

We present a spatio–temporal prediction model that allows forecasting of the criminal activity behavior in a particular region by using supervised classification. The degree of membership of each pattern is interpreted as the forecasted increase or decrease in the criminal activity for the specified time and location. The proposed forecasting model (CR–Ω+) is based on the family of Kora–Q Logical–Combinatorial algorithms operating on large data volumes from several heterogeneous sources using an inductive learning process. We propose several modifications to the original algorithms by Bongard and Baskakova and Zhuravlëv which improve the prediction performance on the studied dataset of criminal activity. We perform two analyses: punctual prediction and tendency analysis, which show that it is possible to predict punctually one of four crimes to be perpetrated (crime family, in a specific space and time), and 66% of effectiveness in the prediction of the place of crime, despite of the noise of the dataset. The tendency analysis yielded an STRMSE (Spatio–Temporal RMSE) of less than 1.0.

Keywords: Logical–combinatorial pattern recognition, forecasting models, supervised classification, crime analysis.

 

RESUMEN

Presentamos un modelo de predicción espacio–temporal que permite la predicción de la actividad criminal dentro de la región estudiada usando clasificación supervisada. El grado de pertenencia de cada patrón es interpretado como el incremento o decremento previsto de la actividad criminal para un tiempo y lugar específico. El modelo propuesto de predicción CR–Ω+ está basado en la familia de los algoritmos Lógico–Combinatorios Kora–Ω para clasificación supervisada. Estos operan sobre volúmenes grandes de datos obtenidos a partir de fuentes heterogéneas de información, con un proceso inductivo de aprendizaje. Proponemos diversas modificaciones al algoritmo original de Bongard, así como el de Baskakova y Zhuravlëv, las cuales mejoran el desempeño de la predicción en el conjunto de datos estudiados de actividad criminal. Realizamos dos análisis: predicción puntual y análisis de tendencias, los cuales muestran que es posible predecir puntualmente uno de cuatro crímenes a ser perpetrados (por familia del crimen) en un tiempo y espacio específicos, así como un 66% de predicción del lugar del crimen, a pesar del ruido en el conjunto de datos de entrada. El análisis de tendencias dio como resultado un RMSE Espacio–Temporal (STRMSE) menor a 1.0.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

Agradecimientos

We thank the support of the Mexican Government (SNI, SIP–IPN, COFAA–IPN, and PIFI–IPN), CONACYT, the Japanese Government and the President of the Municipality of Cuautitlan Izcalli, David Ulises Guzmán Palma, for their support and help in obtaining the data used in this research. The second author is currently a JSPS fellow.

 

References

[1] Amir, M., Patterns in forcible rape. Chicago: University of Chicago Press, 1971.         [ Links ]

[2] Baldwin, J., Bottoms, A. The urban criminal: A study in Sheffield. London: Tavistock Publications, 1976.         [ Links ]

[3] Baskakova, L.V., Y. I. Zuravlëv. Recognition Algorithm Models with Representative Sets and Supporting Sets Systems (in Russian), Zh. Vichislitielnoi Matematiki i Matematicheskoi Fiziki. Tom 21, No.5. URSS, 1981.         [ Links ]

[4] Block, C., STAC hot–spot areas: A statistical tool for law enforcement decisions. In Block, C. R., Dabdoub, M., & Fregly, S. (Eds.), Crime analysis through computer mapping. Washington, DC: Police Executive Research Forum, p. 15–32, 1995.         [ Links ]

[5] Bongard, M.N., et al, Solving geological problems using recognition programs. Journal of Sovietic Geology, No.6, 1963.         [ Links ]

[6] Capone, D., Nichols, W., Urban structure and criminal mobility. American Behavioral Scientist, 20, 199–213, 1976.         [ Links ]

[7] Carrasco–Ochoa, J.A., Representative–Sets based Classifiers, Masters Thesis, CINVESTAV–IPN, Mexico, 1994.         [ Links ]

[8] Cheremesina, N. E., J. Ruiz Shulcloper, Cuestiones metodológicas de la aplicación de modelos matemáticos de Reconocimiento de Patrones en zonas del conocimiento poco formalizadas, Revista Ciencias Matemáticas, XIII(2), La Habana, Cuba, 1992.         [ Links ]

[9] Cressie, N. A. C., Statistics for spatial data. Wiley, New York, 1993.         [ Links ]

[10] De la Vega Doria, L., Extension to the fuzzy case of the KORA–3 algorithm (in Spanish). Masters Thesis, CINVESTAV–IPN, Mexico, 1994.         [ Links ]

[11] De–la–Vega Doria, L., J. Ruiz Shulcloper, Carrasco Ochoa. Fuzzy KORA–Ω Algorithm. Procedings of the 6th European Congress on Inteligent Techniques and Soft Computing, EUFIT, Aachen, Germany, 1998.         [ Links ]

[12] Diukova, E.V., On a parametric model of KORA based recognition algorithms (in Russian), Soovshenia po prikladmoi matematiki, Russia, 1988.         [ Links ]

[13] Goldfarb, L. A new approach in Pattern Recognition, Progress in Machine Intelligence & Pattern Recognition, Ed. L. Kanal, A. Rosenfeld. vol. II, 1985.         [ Links ]

[14] Hirschfield, A., K. Bowers, Mapping and Analysing Crime Data: Lessons from research and practice, Taylor & Francis Inc., 2001.         [ Links ]

[15] Ivaha, C., Al–Madfai, H., Higgs, G., Ware, A. and Corcoran, J., The simple spatial disaggregation approach to spatio–temporal crime forecasting. International Journal of Innovative Computing Information And Control, 3 3: 509–523, 2007.         [ Links ]

[16] Jefferis, E., A multi–method exploration of crime hot spots. Presentation at the Annual Meeting of the Academy of Criminal Justice Sciences, Albuquerque, NM, March 10–14, 1998.         [ Links ]

[17] LeBeau, J. L., The journey to rape: Geographic distance and the rapist's methods of approaching the victim. Journal of Police Science and Administration, 15, 129–136, 1987.         [ Links ]

[18] Leipnik, M., D.P. Albert, GIS in Law Enforcement: Implementation issues and case studies, Taylor & Francis Inc., 2003.         [ Links ]

[19] Levine, N., "Hot Spot" analysis using CrimeStat kernel density interpolation. Presentation at the Annual Meeting of the Academy of Criminal Justice Sciences, Albuquerque, NM, March 10 –14, 1998.         [ Links ]

[20] López–Reyes, N. Ruiz Shulcloper, et al., Un sistema para el pronóstico a corto plazo de tormentas ionosféricas, Reporte de investigación ICIMAF–ACC, 76, La Habana, CUBA, 1981.         [ Links ]

[21] Martínez Trinidad, J.F., A. Guzmán Arenas, The Logical–Combinatorial approach to Pattern Recognition, an overview through selected works, Pattern Recognition 34:4, 2001.         [ Links ]

[22] Mena, J., Investigative Data Mining for Security and Criminal Detection, Butterworth–Heinemann, 2003.         [ Links ]

[23] Molumby, T., Patterns of crime in a university housing project. American Behavioral Scientist, 20, 247259, 1976.         [ Links ]

[24] Newman, O., Defensible space: Crime prevention through urban design. New York, Macmillan, 1973.         [ Links ]

[25] Repetto, T. A., Residential crime. Cambridge, MA, Ballinger, 1972.         [ Links ]

[26] Rossmo, D. K., Target patterns of serial murders: A methodological model. American Journal of Criminal Justice, 17(2), 1–21, 1993.         [ Links ]

[27] Rossmo, D. K., Targeting victims: Serial killers and the urban environment. In O'Reilly–Flemming, T. (Ed.), Serial and mass murder: Theory, research, and policy. Toronto: Canadian Scholars Press, 1996.         [ Links ]

[28] Ruiz Shulcloper J., Tutorial course: Classification with mixed and incomplete data (in Spanish). VII Iberoamerican Conference on Pattern Recognition, México, 2002.         [ Links ]

[29] Ruiz Shulcloper, J., A. Guzmán Arenas, J. F. Martínez Trinidad, Enfoque Lógico–Combinatorio al Reconocimiento de Patrones I: Selección de variables y Clasificación Supervisada. CIC–IPN. Colección de Ciencia de la Computación, 1999.         [ Links ]

[30] Scarr, H. A., Patterns in burglary. 2nd ed., Washington, DC: U.S. Department of Justice, 1973.         [ Links ]