A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES
Accurate crime prediction plays an important role in public safety, providing technical guidance and decision support for the police and government departments. Due to the dynamics and imbalance of crime distribution, it is difficult to build predictive models for it. Specifically, the fine-grained...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2022-10-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLVIII-3-W2-2022/35/2022/isprs-archives-XLVIII-3-W2-2022-35-2022.pdf |
| _version_ | 1811257431602233344 |
|---|---|
| author | X. Lv C. Jing Y. Wang S. Jin |
| author_facet | X. Lv C. Jing Y. Wang S. Jin |
| author_sort | X. Lv |
| collection | DOAJ |
| description | Accurate crime prediction plays an important role in public safety, providing technical guidance and decision support for the police and government departments. Due to the dynamics and imbalance of crime distribution, it is difficult to build predictive models for it. Specifically, the fine-grained and non-linear spatiotemporal dependencies of crime data cannot be captured accurately. In this paper, a neural network model ST-ACLCrime based on ConvLSTM and SE block was proposed to predict the number of theft crimes in hotspot areas. By overlaying ConvLSTM layers, fine-grained spatiotemporal dependencies are captured while preserving spatial location information. To further enhance the global channel feature representation, SE block is used to recalibrate the channel features and enhance the channel inter-dependencies. In addition, the closeness and the period components are set to dynamically capture the dependence of different time trends. We choose the city of Chicago as the study case, and use a multi-level spatial grid to divide the whole city area. The experimental results show that the proposed model exceeds all baseline model, such as HA, CNN, LSTM, CNN-LSTM and ConvLSTM. It was effectively capturing spatiotemporal dependence and improving prediction accuracy. |
| first_indexed | 2024-04-12T17:56:28Z |
| format | Article |
| id | doaj.art-429c4db5681c46e7b42452c6a413bff1 |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-04-12T17:56:28Z |
| publishDate | 2022-10-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-429c4db5681c46e7b42452c6a413bff12022-12-22T03:22:19ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342022-10-01XLVIII-3-W2-2022354110.5194/isprs-archives-XLVIII-3-W2-2022-35-2022A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMESX. Lv0C. Jing1Y. Wang2S. Jin3School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaSchool of Information Engineering, China University of Geosciences, Beijing 10083, ChinaSchool of Earth and Space Sciences, Peking University, Beijing 100871, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 100044, ChinaAccurate crime prediction plays an important role in public safety, providing technical guidance and decision support for the police and government departments. Due to the dynamics and imbalance of crime distribution, it is difficult to build predictive models for it. Specifically, the fine-grained and non-linear spatiotemporal dependencies of crime data cannot be captured accurately. In this paper, a neural network model ST-ACLCrime based on ConvLSTM and SE block was proposed to predict the number of theft crimes in hotspot areas. By overlaying ConvLSTM layers, fine-grained spatiotemporal dependencies are captured while preserving spatial location information. To further enhance the global channel feature representation, SE block is used to recalibrate the channel features and enhance the channel inter-dependencies. In addition, the closeness and the period components are set to dynamically capture the dependence of different time trends. We choose the city of Chicago as the study case, and use a multi-level spatial grid to divide the whole city area. The experimental results show that the proposed model exceeds all baseline model, such as HA, CNN, LSTM, CNN-LSTM and ConvLSTM. It was effectively capturing spatiotemporal dependence and improving prediction accuracy.https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLVIII-3-W2-2022/35/2022/isprs-archives-XLVIII-3-W2-2022-35-2022.pdf |
| spellingShingle | X. Lv C. Jing Y. Wang S. Jin A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES |
| title_full | A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES |
| title_fullStr | A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES |
| title_full_unstemmed | A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES |
| title_short | A DEEP NEURAL NETWORK FOR SPATIOTEMPORAL PREDICTION OF THEFT CRIMES |
| title_sort | deep neural network for spatiotemporal prediction of theft crimes |
| url | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLVIII-3-W2-2022/35/2022/isprs-archives-XLVIII-3-W2-2022-35-2022.pdf |
| work_keys_str_mv | AT xlv adeepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT cjing adeepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT ywang adeepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT sjin adeepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT xlv deepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT cjing deepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT ywang deepneuralnetworkforspatiotemporalpredictionoftheftcrimes AT sjin deepneuralnetworkforspatiotemporalpredictionoftheftcrimes |