CRIME ACTIVITIES TRENDS ANALYSIS WITH A MACHINE LEARNING TECHNIQUE

Author Reg. ID: TIJER_109242

Published Paper Id: TIJER2309159

Published In: Volume 10 Issue 9, September-2023

Abstract: Criminal activity is a major problem faced by societies worldwide. Identifying factors and patterns that contribute to criminal activity can help law enforcement agencies to prevent and reduce crime. The study purpose was to identify the trends in the rate of criminal activities over a given period of time and proposed a model for the data. It has been noted that both ARIMA(3,0,0) and SARIMA (3,0,0)(1,0,0)12 models are good for the crime data. The model Root Mean Square Error (RMSE) of 2.813 shows that that the predicted values are about 97% close to the actual dataset, indicating a good model. Again, the Mean Absolute Percentage Error (MAPE) of 2.048 means the model prediction accuracy is about 98%. Thus, we are certain that our forecasted values with the model are a true reflection of the nature of the crime rate for some periods ahead. In addition, the result shows an up-rise in the crime rate. This means that criminal activities will continue to rise if proper measures or policies are not implemented to curb them.

Keywords: Criminal, Effective strategies, Criminology, Differencing, Auto Regressive

Downloads: 000113

Page No: b207-b213

Country: Takoradi, Takoradi, Ghana

Research Area: Applied Mathematics

Published Paper URL: https://tijer.org/TIJER/viewpaperforall?paper=TIJER2309159

Published Paper PDF: https://tijer.org/TIJER/papers/TIJER2309159

Click Here to Download This Article

Article Preview