Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model
While the use of crime data has been widely advocated in the literature, its availability is often limited to large urban cities and isolated databases that tend not to allow for spatial comparisons. This paper presents an efficient machine learning framework capable of predicting spatial crime occu...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/9/11/645 |
| _version_ | 1827703301432410112 |
|---|---|
| author | Yasmine Lamari Bartol Freskura Anass Abdessamad Sarah Eichberg Simon de Bonviller |
| author_facet | Yasmine Lamari Bartol Freskura Anass Abdessamad Sarah Eichberg Simon de Bonviller |
| author_sort | Yasmine Lamari |
| collection | DOAJ |
| description | While the use of crime data has been widely advocated in the literature, its availability is often limited to large urban cities and isolated databases that tend not to allow for spatial comparisons. This paper presents an efficient machine learning framework capable of predicting spatial crime occurrences, without using past crime as a predictor, and at a relatively high resolution: the U.S. Census Block Group level. The proposed framework is based on an in-depth multidisciplinary literature review allowing the selection of 188 best-fit crime predictors from socio-economic, demographic, spatial, and environmental data. Such data are published periodically for the entire United States. The selection of the appropriate predictive model was made through a comparative study of different machine learning families of algorithms, including generalized linear models, deep learning, and ensemble learning. The gradient boosting model was found to yield the most accurate predictions for violent crimes, property crimes, motor vehicle thefts, vandalism, and the total count of crimes. Extensive experiments on real-world datasets of crimes reported in 11 U.S. cities demonstrated that the proposed framework achieves an accuracy of 73% and 77% when predicting property crimes and violent crimes, respectively. |
| first_indexed | 2024-03-10T15:15:14Z |
| format | Article |
| id | doaj.art-df6f49e9154d410fafbf2378ad134883 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-10T15:15:14Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-df6f49e9154d410fafbf2378ad1348832023-11-20T19:02:15ZengMDPI AGISPRS International Journal of Geo-Information2220-99642020-10-0191164510.3390/ijgi9110645Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning ModelYasmine Lamari0Bartol Freskura1Anass Abdessamad2Sarah Eichberg3Simon de Bonviller4Augurisk, Inc., Wilmington, DE 19802, USAVelebit Artificial Intelligence LLC, 10000 Zagreb, CroatiaAugurisk, Inc., Wilmington, DE 19802, USAIndependent Researcher, Dunedin, FL 34698, USAAugurisk, Inc., Wilmington, DE 19802, USAWhile the use of crime data has been widely advocated in the literature, its availability is often limited to large urban cities and isolated databases that tend not to allow for spatial comparisons. This paper presents an efficient machine learning framework capable of predicting spatial crime occurrences, without using past crime as a predictor, and at a relatively high resolution: the U.S. Census Block Group level. The proposed framework is based on an in-depth multidisciplinary literature review allowing the selection of 188 best-fit crime predictors from socio-economic, demographic, spatial, and environmental data. Such data are published periodically for the entire United States. The selection of the appropriate predictive model was made through a comparative study of different machine learning families of algorithms, including generalized linear models, deep learning, and ensemble learning. The gradient boosting model was found to yield the most accurate predictions for violent crimes, property crimes, motor vehicle thefts, vandalism, and the total count of crimes. Extensive experiments on real-world datasets of crimes reported in 11 U.S. cities demonstrated that the proposed framework achieves an accuracy of 73% and 77% when predicting property crimes and violent crimes, respectively.https://www.mdpi.com/2220-9964/9/11/645crime predictionensemble learningmachine learningregression |
| spellingShingle | Yasmine Lamari Bartol Freskura Anass Abdessamad Sarah Eichberg Simon de Bonviller Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model ISPRS International Journal of Geo-Information crime prediction ensemble learning machine learning regression |
| title | Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model |
| title_full | Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model |
| title_fullStr | Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model |
| title_full_unstemmed | Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model |
| title_short | Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model |
| title_sort | predicting spatial crime occurrences through an efficient ensemble learning model |
| topic | crime prediction ensemble learning machine learning regression |
| url | https://www.mdpi.com/2220-9964/9/11/645 |
| work_keys_str_mv | AT yasminelamari predictingspatialcrimeoccurrencesthroughanefficientensemblelearningmodel AT bartolfreskura predictingspatialcrimeoccurrencesthroughanefficientensemblelearningmodel AT anassabdessamad predictingspatialcrimeoccurrencesthroughanefficientensemblelearningmodel AT saraheichberg predictingspatialcrimeoccurrencesthroughanefficientensemblelearningmodel AT simondebonviller predictingspatialcrimeoccurrencesthroughanefficientensemblelearningmodel |