Unstructured Malay Text Analytics Model in Crime

Abstract

The concern in crime prevention to realize a safe environment for public's welfare is one of the central agenda to government authorities. The increasing volume of crime reports with unstructured data that are heavily loaded with text presents the biggest challenge to human experts in analysing the data efficiently as it contains lots of hidden and valuable information such as modus operandi and common crime patterns. The capacity limitations in human information processing demand an effective data analytics model to uncover these hidden and potentially meaningful information from large-scale of data especially in Malay language for improving analysis and supporting decision-making in crime investigation. The objective of this paper is to propose Unstructured Malay Text Analytics Model to support serial crime investigation based on modus operandi patterns. The significance output of this paper is to assist experts especially Royal Malaysia Police in crime investigation based on modus operandi patterns analysis as one of the efforts in reducing rate of crime index.