Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact

Repository Web            Google
View ALL Data Sets

UrbanGB, urban road accidents coordinates labelled by the urban center Data Set
Download: Data Folder, Data Set Description

Abstract: Coordinates (longitude and latitude) of 360177 road accidents occurred in urban areas in Great Britain, and labelled according to the urban center where they occurred (469 possible labels).

Data Set Characteristics:  


Number of Instances:




Attribute Characteristics:


Number of Attributes:


Date Donated


Associated Tasks:


Missing Values?


Number of Web Hits:



Coordinates data obtained by filtering the dataset at
Raw data originally available from, released under the Open Government Licence (
Filtering and labelling performed by Carlo Baldassi, Bocconi University (carlo.baldassi '@', licenced under the Open Database License (

Data Set Information:

Please see the included file for details on the contents, format and purpose of the dataset, and the procedure used to create it.

Attribute Information:

The input data contains geographical coordinates in the ranges [-5.55599, 1.75834] (longitude) and [50.0797, 57.6956] (latitude). It is advisable to scale the longitude down by a factor of 1.7 for the purpose of geographical clustering.
The labels are provided in a separate file (one label per entry, ranging from 1 to 469).
The centroids (barycenters of each partition) are also provided, in a separate file.
See the included file for further details.

Relevant Papers:

C. Baldassi, 'Recombinator-k-means: A population based algorithm that exploits k-means++ for recombination', [Web Link], 2019

Citation Request:

Please refer to the Machine Learning Repository's citation policy

Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML