Developing novel spatial statistics in population health has been researched for many years in the Institute of Geographic Sciences and Natural Resources Research (IGSNRR). Recently, a research group led by Prof. WANG Jinfeng published a paper “Using rough set theory to identify villages affected by birth defects: the example of Heshun, Shanxi, China” in International Journal of Geographical Information Science.

Researchers use rough set theory to explore spatial decision rules in neural-tube birth defects and searches for novel spatial factors related to the disease. Rough set theory is data-driven, and the advantage of this approach is that it does not require the user to make any a priori assumptions about the data. It can be used to find attributes efficiently, a critical step in the decision-making process, through the computing of reducts. Furthermore, the decision rule generated can be used to predict or classify unseen objects. From the accuracy assessment of the identification through an error matrix, it can be seen that the spatial rules synthesized by the rough set have reasonably good generalization. On the other hand, as a preliminary step to finding the direct causes of the disease, many useful conclusions can be drawn from the rules generated. For example, the rules found in our experiment reveal that the watershed and gradient are important. Then further in situ inspections and experiments of the area need to be considered in order to identify the real causes that hide behind the simple relationship.