Uncover Hidden Patterns in Crime Data Using Frequent Pattern-Growth Algorithm

Chew, Yu Ming (2024) Uncover Hidden Patterns in Crime Data Using Frequent Pattern-Growth Algorithm. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RDS_Chew Yu Ming_Fulltext.pdf Restricted to Registered users only Download (1MB)

Abstract

The intensifying concern over rising crime rates, particularly in urban areas like Chicago, underscores the imperative to comprehensively analyze crime data for insights crucial to public safety enhancement. This study concentrates on Chicago’s extensive crime dataset from 2022, utilizing association rules mining, notably the Frequent Pattern-Growth algorithm, to uncover intricate relationships and latent connections within the data. By meticulously employing exploratory data analysis and visualization techniques, this research aims to predict potential outcomes stemming from specific events, unravel obscured patterns, and discern profound relationships between crime occurrences, temporal aspects, and spatial distribution. The objectives entail leveraging association rules to delve beyond surface-level insights, revealing nuanced connections that traditional statistical methods often fail to capture. This approach is poised to illuminate nuanced correlations between diverse crime types, locations, and socio-economic factors, potentially shedding light on underlying causes and behavioral patterns of criminal activities. By advocating for the superiority of association rules mining in deciphering complex datasets, this project seeks to significantly contribute to the field of crime analysis. The anticipated outcomes hold promise in informing policy decisions and law enforcement strategies, ultimately fostering a safer urban environment by harnessing the power of comprehensive crime data analyzes to proactively address criminal trends and their root causes