PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 3, 2026InformationOpen Access

Enhancing Multi-Level Spatio-Temporal Forecasting of Adjudicated Crime Occurrence Trends in Indonesia

View Full Paper
Ask AI
Bookmark
Share

Authors

FAFirman ArifmanTMTeddy MantoroMAMedia Anugerah Ayu

Discussion

Loading...

Member takes

Overview

Machine learning improves crime occurrence predictions in Indonesia, suggesting better judicial resource allocation.

Key Points

  • The aim is to develop a machine learning framework for predicting crime trends in Indonesia's judiciary system.
  • Developed a multi-level spatio-temporal forecasting framework using machine learning.
  • Analyzed 95,666 adjudicated crime records from January 2023 to June 2024.
  • Employed CRISP-DM methodology to train hybrid STL-XGBoost model.
  • Conducted DBSCAN spatial clustering to identify crime hubs.
  • Achieved an R2 of 0.8070, MAE of 16.52, and sMAPE of 9.76% for forecasting.
  • Identified 29 distinct adjudicated crime hubs concentrated along urban corridors.
  • Reported systematic judicial funnel attrition, indicating large discrepancies between reported and adjudicated crimes.

Cite This Study

Arifman et al. (2026) studied this question.

synapsesocial.com/papers/69cf5f225a333a821460e00dhttps://doi.org/10.3390/info17040331
View Full Paper
Ask AI
Bookmark
Share