This project is a comprehensive Chicago Crime Analysis study that I conducted to investigate spatial, temporal, and socioeconomic patterns within the city’s crime landscape. Using publicly available datasets from the Chicago Data Portal, I designed and executed a full research workflow—beginning with data acquisition and cleaning, followed by exploratory analysis, statistical modelling, and the development of interpretable visualisations to uncover underlying trends.

The report documents the construction of a rigorous analytical pipeline primarily using Python and its various machine learning libraries, incorporating geospatial mapping, regression techniques, clustering algorithms, and time-series decomposition. I examined how crime rates vary across neighbourhoods, how they evolve over time, and what socioeconomic variables correlate meaningfully with these variations. Each methodological choice is supported by a clear rationale, with careful attention to issues such as missing data, reporting inconsistencies, and model reliability.

Beyond identifying patterns, the study evaluates their implications for urban policy, resource allocation, and community-level risk assessment. I also reflect critically on the limitations of predictive modelling in criminology and suggest directions for more robust future research.

This Chicago Crime Analysis project was formally recognised for its analytical depth, technical execution, and originality—earning me the CREST Gold Award from the British Science Association for independent scientific research.