Identifying key influential parameters of high profile criminals through statistical correlation

• Published in: Cluster computing Vol. 22; no. Suppl 3; pp. 7135 - 7148
• Main Authors: Wadood, Danish, Rauf, Azhar, Khusro, Shah, Ali, Shaukat
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.05.2019; Springer Nature B.V
• Subjects: Computer Communication Networks; Computer Science; Correlation analysis; Court hearings & proceedings; Crime; Criminal investigations; Criminal statistics; Criminology; Data mining; Fear of crime; Honor killings; Law enforcement; Operating Systems; Parameter identification; Prisoners; Prisons; Processor Architectures; Social networks; Statistical correlation; Uniform Crime Reports; Influential parameters; Correlation; Data mining; Criminology
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-017-1059-1

Identification of influential parameters using data mining and statistical tools is an important subject of criminology. These parameters are further used in criminal prediction and crime reduction which is a well-known problem in law enforcement agencies. In this research, we have analyzed the 5 years data of the Prison Department of the Government of KP. More than thirty parameters related to criminals were cleaned, preprocessed and analyzed. Correlation of these parameters with high act crimes was identified. There are five parameters which have strong correlation with high act crimes that includes total number of group members, recovery of assets, total number of hearings, crime frequency, and total number of non-blood visitors. On the other hand, there are three parameters that have negative correlation with high act crimes which includes education level of a criminal, total number of dependents, and prison duration. The overall objective of this research is to identify influential parameters related to high category crimes and support the law enforcement agencies of the KP province in reducing the crime ratio.