Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Innovative LSGTime Model for Crime Spatiotemporal Prediction Based on MindSpore Framework

Zhenkai Qin, BaoZhong Wei, Caifeng Gao

TL;DR

The paper tackles spatiotemporal crime forecasting in urban settings by introducing LGSTime, a hybrid architecture that fuses LSTM, GRU, and Multi-head Sparse Self-attention to capture long-term temporal patterns and complex spatial interactions. By leveraging LSTM for long-range dependencies, GRU for short-term dynamics, and MHSA for cross-feature spatiotemporal relationships with sparsity for efficiency, the model achieves state-of-the-art results on four real-world crime datasets. Ablation studies demonstrate the added value of the MHSA component, while experiments confirm improvements in MSE, MAE, and RMSE over baselines. The work has practical implications for resource allocation and public safety, and suggests future directions in multimodal data fusion and cross-domain deployment using MindSpore.

Abstract

With the acceleration of urbanization, the spatiotemporal characteristics of criminal activities have become increasingly complex. Accurate prediction of crime distribution is crucial for optimizing the allocation of police resources and preventing crime. This paper proposes LGSTime, a crime spatiotemporal prediction model that integrates Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and the Multi-head Sparse Self-attention mechanism. LSTM and GRU capture long-term dependencies in crime time series, such as seasonality and periodicity, through their unique gating mechanisms. The Multi-head Sparse Self-attention mechanism, on the other hand, focuses on both temporal and spatial features of criminal events simultaneously through parallel processing and sparsification techniques, significantly improving computational efficiency and prediction accuracy. The integrated model leverages the strengths of each technique to better handle complex spatiotemporal data. Experimental findings demonstrate that the model attains optimal performance across four real - world crime datasets. In comparison to the CNN model, it exhibits performance enhancements of 2.8\%, 1.9\%, and 1.4\% in the Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) metrics respectively. These results offer a valuable reference for tackling the challenges in crime prediction.

Innovative LSGTime Model for Crime Spatiotemporal Prediction Based on MindSpore Framework

TL;DR

The paper tackles spatiotemporal crime forecasting in urban settings by introducing LGSTime, a hybrid architecture that fuses LSTM, GRU, and Multi-head Sparse Self-attention to capture long-term temporal patterns and complex spatial interactions. By leveraging LSTM for long-range dependencies, GRU for short-term dynamics, and MHSA for cross-feature spatiotemporal relationships with sparsity for efficiency, the model achieves state-of-the-art results on four real-world crime datasets. Ablation studies demonstrate the added value of the MHSA component, while experiments confirm improvements in MSE, MAE, and RMSE over baselines. The work has practical implications for resource allocation and public safety, and suggests future directions in multimodal data fusion and cross-domain deployment using MindSpore.

Abstract

With the acceleration of urbanization, the spatiotemporal characteristics of criminal activities have become increasingly complex. Accurate prediction of crime distribution is crucial for optimizing the allocation of police resources and preventing crime. This paper proposes LGSTime, a crime spatiotemporal prediction model that integrates Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and the Multi-head Sparse Self-attention mechanism. LSTM and GRU capture long-term dependencies in crime time series, such as seasonality and periodicity, through their unique gating mechanisms. The Multi-head Sparse Self-attention mechanism, on the other hand, focuses on both temporal and spatial features of criminal events simultaneously through parallel processing and sparsification techniques, significantly improving computational efficiency and prediction accuracy. The integrated model leverages the strengths of each technique to better handle complex spatiotemporal data. Experimental findings demonstrate that the model attains optimal performance across four real - world crime datasets. In comparison to the CNN model, it exhibits performance enhancements of 2.8\%, 1.9\%, and 1.4\% in the Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) metrics respectively. These results offer a valuable reference for tackling the challenges in crime prediction.

Paper Structure

This paper contains 15 sections, 20 equations, 1 figure, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. LSTM In Crime Prediction
  4. GRU In Crime Prediction
  5. MHSA In Crime Prediction
  6. Methodology
  7. LSTM Gate Mechanism
  8. GRU Gate Mechanism
  9. Multi-head Sparse Self-attention
  10. Experiments
  11. Datasets
  12. Implementation Details
  13. Multivariate Result
  14. Ablation Research
  15. Conclusions

Figures (1)

  • Figure 1: The LGSTime model consists of three modules: LSTM, capable of capturing long-term dependencies in time series data; GRU, focused on short-term dependencies; and a Multi-head Sparse self-attention module that identifies complex relationships between different features within the input sequence.