SASB: Spatial Activity Summarization using constant area Buffers

Given a spatial network, a spatial neighbourhood, a collection of activities, a constant buffer area and a number of clusters to identify (k), the problem of spatial activity summarization using constant area buffers is to choose the user defined k constant area buffers which maximizes the number of activities covered. Spatial activity summarization is an important societal problem and has been used by authorities to curb crime [16, 7, 13] and respond to natural disasters [18, 5, 1]. Oliver et al. [18] proved, in specific cases, network based summarization is a better approach than geometry based ones whereas the existence of geometry based approaches for the last 2 decades [10] proves their acceptance with the environmental criminologists and the authorities [7]. This paper presents an alternative way of summarizing spatial activities which helps the user summarize activities using k different buffers. The problem is of choosing the k best buffers out the available β buffers which provides us with a β choose k ( β k ) ways of accomplishing it. We use a greedy approach to maximize the number of activities covered in the k buffers. First principle examples show that this approach’s performance is comparable to the the above-mentioned approaches and in most instances presented in the examples it outperforms both.