Using Big Data Analytics to Identify Trends and Group Crimes through Clustering

The incidence of crime in a city presents a challenge in the absence of trend analysis that impacts citizen security. The objective of this research was to analyze and visualize crime trends in the area, using the concepts and fundamentals of Big Data Analytics, Data Mining and Clustering, the problem is addressed with a quantitative approach, using the CRISP-DM process, Principal Component Analysis (PCA) and the K-Means algorithm for clustering. Validation is performed with the Elbow Score and the Average Silhouette method, ensuring the robustness of the data clustering. The results show that crimes against property, such as robbery and theft, are the most frequent. Four crime clusters are identified, each associated with a specific category, providing a detailed view of crime distribution. Comparison with previous studies highlights the effectiveness of Big Data technologies in reducing crime, providing a solid basis for more accurate security strategies.