Mixture model based time-series clustering of crime data across spatial entities
A crime analysis system, method, and apparatus comprising at least one processor and a storage device communicatively coupled to the at least one processor, the storage device storing instructions which, when executed by the at least one processor, cause the processor to perform operations comprising receiving information provided by one or more data collection source, storing the information, wherein the stored information is formatted, processing the information to generate crime clustering data associated with at least one region and at least one crime, processing the crime clustering data associated with at least one region and at least one crime to generate benchmarking of the at least one region with at least one other region, and providing crime clustering data associated with at least one region and at least one crime, and benchmarking of the at least one region with at least one other region for presentation through a user interface.
1. A crime analysis system comprising:
at least one processor; and
a storage device communicatively coupled to said at least one processor, said storage device storing instructions which, when executed by said at least one processor, cause said at least one processor to perform operations comprising:
receiving information provided by at least one data collection source;
storing said information, wherein said stored information is formatted;
processing said information to generate crime clustering data associated with at least one region and at least one crime, wherein said processing of said information to generate crime clustering data associated with said at least one region and said at least one crime further comprises:
extracting features from said information;
identifying feature distribution associated with said extracted features; and
clustering said information according to feature extraction and feature distribution, wherein said clustering of said information further comprises: applying a mixture model based time-series clustering framework to said identified extracted features, wherein said mixture model based time-series clustering framework uses a set of statistical and domain-level features for clustering across spatial entities, and wherein said mixture model based time-series clustering framework handles heterogeneity in time series-data from a scheduled job according to spatial parameters and crime types;
processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region;
providing crime clustering data associated with at least one region and at least one crime, and benchmarking of said at least one region with at least one other region for presentation through a user interface; and
permitting a granulating of said at least one region and said at least one other region to be adjustable according to a user preference.
2. The system of claim 1 , wherein said information comprises at least one of: census data; region specific crime data; GIS data; and spatial data.
3. The system of claim 1 , wherein extracting features from said information further comprises: calculating an average Maharaj distance associated with said information.
4. The system of claim 1 , wherein processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region further comprises: calculating an L2 Norm between said at least one region and said at least one other region.
5. The system of claim 1 , wherein providing crime clustering data associated with at least one region and at least one crime, and benchmarking of said at least one region with at least one other region for presentation through a user interface further comprises:
identifying patterns in crime related statistics for said at least one region over a given time; and
identifying best crime prevention practices according to said benchmarking of said at least one region with said at least one other region.
6. The system of claim 1 , wherein said formatted information comprises time series data of at least one crime type in at least one region.
7. An apparatus comprising non-transitory computer readable storage media storing instructions which, when executed by at least one processor, cause the at least one processor to perform operations comprising:
receiving information provided by at least one data collection source;
storing said information, wherein said stored information is formatted;
processing said information to generate crime clustering data associated with at least one region and at least one crime, wherein said processing of said information to generate crime clustering data associated with said at least one region and said at least one crime further comprises:
extracting features from said information;
identifying feature distribution associated with said extracted features; and
clustering said information according to feature extraction and feature distribution, wherein said clustering of said information further comprises: applying a mixture model based time-series clustering framework to said identified extracted features, wherein said mixture model based time-series clustering framework uses a set of statistical and domain-level features for clustering across spatial entities, and wherein said mixture model based time-series clustering framework handles heterogeneity in time series-data from a scheduled job according to spatial parameters and crime types;
processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region;
providing crime clustering data associated with at least one region and at least one crime, and benchmarking of said at least one region with at least one other region for presentation through a user interface; and
permitting a granulating of said at least one region and said at least one other region to be adjustable according to a user preference.
8. The apparatus of claim 7 , wherein processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region further comprises:
calculating an L2 Norm between said at least one region and said at least one other region.
9. A computer-implemented method comprising:
receiving information provided by at least one data collection source;
storing said information, wherein said stored information is formatted;
processing said information to generate crime clustering data associated with at least one region and at least one crime, wherein said processing of said information to generate crime clustering data associated with said at least one region and said at least one crime further comprises:
extracting features from said information;
identifying feature distribution associated with said extracted features; and
clustering said information according to feature extraction and feature distribution, wherein said clustering of said information further comprises: applying a mixture model based time-series clustering framework to said identified extracted features, wherein said mixture model based time-series clustering framework uses a set of statistical and domain-level features for clustering across spatial entities, and wherein said mixture model based time-series clustering framework handles heterogeneity in time series-data from a scheduled job according to spatial parameters and crime types;
processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region;
providing crime clustering data associated with at least one region and at least one crime, and benchmarking of said at least one region with at least one other region for presentation through a user interface; and
permitting a granulating of said at least one region and said at least one other region to be adjustable according to a user preference.
10. The method of claim 9 , wherein said information comprises at least one of: census data; region specific crime data; GIS data; and spatial data.
11. The method of claim 9 , wherein extracting features from said information further comprises: calculating an average Maharaj distance associated with said information.
12. The method of claim 9 , wherein processing said crime clustering data associated with at least one region and at least one crime to generate benchmarking of said at least one region with at least one other region further comprises: calculating an L2 Norm between said at least one region and said at least one other region.
13. The method of claim 9 , wherein providing crime clustering data associated with at least one region and at least one crime, and benchmarking of said at least one region with at least one other region for presentation through a user interface further comprises:
identifying patterns in crime related statistics for said at least one region over a given time; and
identifying best crime prevention practices according to said benchmarking of said at least one region with said at least one other region.
14. The method of claim 9 , wherein said formatted information comprises time series data of at least one crime type in at least one region.