Based on the introduction of the method of constructive learning artificial neural networks(ANN) and the theory of rough sets,the model of the macroeconomic early-warning system is built which is based on the rough set and the constructive learning ANN based on the covering algorithm.

Analysis of data about economic eariy-warning based on rough sets theory is processed and the mining of the hidden knowledge,the acquisition of the decision rules,the reduction of the decision rules is processed and the minimal decision rules are deduced,A new way for the knowledge acquisition of economic early-warning is supplied.

If the fluctuation goes beyond a certain range, it will damage the development of economy;therefore, the research of economic early-warning is needed.

Design and implementation of industry economy early warning system on data warehouse;

On the basis of theoretical analysis and real practice,a quantitative method for economy early warning indicator screening-Applied contribution analysis,is put forward.

Research on economic early warning based on maximum entropy;

As Economic early warning is the cognition and judgment of economic status, its results affect the choice of macroeconomic policies, it has been one of the most important research fields of economics, and widely concerned about by all governments and public.

Research on economic warning mechanism of gold enterprises;

Secondly, according to the selection principles of pre-warning index and the impact of tailing disasters on downstream people and downstream property, the regional alert index system for tailing reservoir is established, which includes 9 indexes, such as the risk grade of tailing reservoir, topographical slope, geological s.