First, an inventory-based training set is adopted to obtain some statistical parameters required to build a non-settlement mask.
Two commonly adopted allocations, i.e., the non-settlement profit allocation and the Shapley-value based profit allocation, are analyzed and compared.
The percent share of other reported risk factors such as cancellation and non-settlement of marriage, love affairs, drug abuse/addiction, failure in examination, bankruptcy, unemployment, poverty have a relatively lesser share than these two major risk factors.
Our results show that the non-settlement allocation is not in the core and does not preserve fairness, and the Shapley-value based profit allocation is in the core and demonstrates fairness.
Hence, if non-settlement of particles is confirmed, then particle dispersion would be a better technique than other enhancement techniques i.e. improving heat exchanger design (using fins) and inserting non-moving high conductivity structures as these techniques would significantly reduce the volume of PCM in the system.
An equal sized set of 'non-settlement' grid squares were randomly selected from non-settlement Africover classes to test whether the predicted settlement maps had identified false areas of settlement [13].
The highest overall accuracy (combined settlement and non-settlement accuracies) was 85.5% for Kenya, but results were similar across the region.
Within these polygons, image classes were highlighted and merged iteratively to best represent the training polygon extents, whilst discounting clear non-settlement land covers within the polygons, and produce a satellite imagery derived settlement map.
In general, the non-settlement test pixels were mapped more accurately than the Africover-defined settlement pixels, although all accuracies were above 70% and Kappa values all above 0.55 (table 1), indicating a good to excellent agreement [21].
Spectral confusion in a few small areas also contributed to potential settlement mapping errors, where what appeared to be non-settlement land cover had an almost identical spectral and radar response to settlements identified using Africover polygons.
Next, by using the ArcGIS® 10 software package, settlement areas were separated from non-settlement land cover.
