A Unified Framework for Crime Prediction Leveraging Contextual and Interaction-Based Feature Engineering DOI Creative Commons

E. Monika,

T. Rajesh Kumar

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Окт. 21, 2024

Abstract The prediction of crime holds significant importance in the realm law enforcement and public safety endeavors. This research paper presents a framework aimed at improving models through integration contextual interaction feature engineering methodologies. study novel methodology that uses minimal spanning trees (MST) within directed graph to depict relationships between incidents specific locations. approach identifies correlations instances criminal activity, enabling creation more intricate forecasting models. suggested framework's effectiveness is assessed by employing diverse classifiers performance metrics, such as accuracy, precision, recall, F1-score. findings indicate technique outperforms current methodologies, highlighting its properly evidence-based decision-making endeavours. with dimensionality reduction graph-based modelling this helps progress approaches.

Язык: Английский

A multi-modal geospatial–temporal LSTM based deep learning framework for predictive modeling of urban mobility patterns DOI Creative Commons

Sangeetha S.K.B,

Sandeep Kumar Mathivanan,

Hariharan Rajadurai

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Дек. 30, 2024

Язык: Английский

Процитировано

3
Crime-associated inequality in geographical access to education: Insights from the municipality of Rio de Janeiro DOI Creative Commons
Steffen Knoblauch,

R. Muthusamy,

Mark Moritz

и другие.

Cities, Год журнала: 2025, Номер 160, С. 105818 - 105818

Опубликована: Фев. 27, 2025

Язык: Английский

Процитировано

0
A Unified Framework for Crime Prediction Leveraging Contextual and Interaction-Based Feature Engineering DOI Creative Commons

E. Monika,

T. Rajesh Kumar

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Окт. 21, 2024

Abstract The prediction of crime holds significant importance in the realm law enforcement and public safety endeavors. This research paper presents a framework aimed at improving models through integration contextual interaction feature engineering methodologies. study novel methodology that uses minimal spanning trees (MST) within directed graph to depict relationships between incidents specific locations. approach identifies correlations instances criminal activity, enabling creation more intricate forecasting models. suggested framework's effectiveness is assessed by employing diverse classifiers performance metrics, such as accuracy, precision, recall, F1-score. findings indicate technique outperforms current methodologies, highlighting its properly evidence-based decision-making endeavours. with dimensionality reduction graph-based modelling this helps progress approaches.

Язык: Английский

Процитировано

0