TY - JOUR
T1 - The CR-ω+ classification algorithm for spatio-temporal prediction of criminal activity
AU - Godoy-Calderón, S.
AU - Calvo, H.
AU - Martínez-Hernández, V. M.
AU - Moreno-Armendáriz, M. A.
PY - 2010/4
Y1 - 2010/4
N2 - 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-ω 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.
AB - 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-ω 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.
KW - Crime analysis
KW - Forecasting models
KW - Logical-combinatorial pattern recognition
KW - Supervised classification
UR - http://www.scopus.com/inward/record.url?scp=84860878320&partnerID=8YFLogxK
M3 - Artículo
SN - 1665-6423
VL - 8
SP - 5
EP - 25
JO - Journal of Applied Research and Technology
JF - Journal of Applied Research and Technology
IS - 1
ER -