CRIME RATE PREDICTION USING THE RANDOM FOREST ALGORITHM

M. Abdulraheem; J. B. Awotunde; I. D. Oladipo; M. O. Adeleke; J. N. Ndunagu; J. A. Ayantola; A. Mohammed

M. Abdulraheem Department of Computer Science, Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, 240003, Kwara State, Nigeria
J. B. Awotunde Department of Computer Science, Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, 240003, Kwara State, Nigeria
I. D. Oladipo Department of Computer Science, Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, 240003, Kwara State, Nigeria
M. O. Adeleke Department of Statistics, Faculty of Physical Sciences, University of Ilorin, Ilorin, 240003, Kwara State, Nigeria.
J. N. Ndunagu Department of Computer Science, National Open University of Nigeria, Abuja, Nigeria
J. A. Ayantola Department of Computer Science, Osun State Polytechnic, Iree Osun State, Nigeria
A. Mohammed Department of Computer Science, Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, 240003, Kwara State, Nigeria

Keywords: Crime rate, Radom Forest, Corruption, Machine learning, Crime network, Kidnapping, Cybercrime

Abstract

An act that creates crimes punishable by law is characterized as a crime. Rape, fraud, terrorism, kidnapping, burglary, murder, and other crimes are common in Nigeria. Examples are cybercrime, bribery and corruption, robbery, money laundering, among other crimes. Crime is a harmful and widespread social issue that affects individuals all around the world. The rate of crime has risen dramatically in recent years. To cut down on crime, at any rate, law enforcements must take preventative actions. To protect society against crime, modern systems and new technologies are required. Although accurate real-time crime study is on aid in reducing crime rates, they are nonetheless useless. As crime occurrences are dependent on, this is a difficult subject for the scientific community to solve. Therefore, this paper proposes machine learning algorithm to indicate the frequency and pattern of crimes based on the data collected and to show the extent of crime in a particular region. Various visualization approaches and machine learning algorithms are used in this study to anticipate the crime distribution over a large area. In the first stage, raw datasets were processed and visualized according to the requirements. Then, to extract knowledge from these massive datasets, machine learning methods were deployed and uncover hidden patterns in the data, which were then utilized to investigate and report on crime patterns, It is beneficial to crime analysts. Investigate these crime networks using a variety of interactive crime visualizations. As a result, it is helpful in crime prevention.