Macroeconomic policy making is a complex systematic process, which requires in-depth understanding of current economic situation, prediction of future economic trend, and proper policy evaluation measurements. Instruments of early warning, and policy simulation are often employed in macroeconomic policy making. However, no matter how well it is developed, any single instrument is often inadequate for policy making support, because of the gap between theories and practice. In this paper, macroeconomic early warning theories are integrated with the policy decision support concepts. Three stages are involved in the macroeconomic policy making process: monitoring, forecasting, and policy simulation. Based on this idea, an integrated alert-response framework is proposed with one corresponding module for each stage. Within this framework, not only information can be exchanged freely among these modules, but the monitoringforecasting-simulation process can run smoothly to realize timeliness and efficient policy making support. Moreover, a knowledge base is incorporated into the framework to support the economic early warning and policy making support process. Therefore, this framework is featured in integration and a final all-round report, including current economic status, future trend prediction, policy making suggestions, external information, and expert opinions, can be generated. An implementation of this framework was developed for China's macroeconomic adjustment and has been put into practice since early 2006. A case of national economic growth analysis based on the proposed framework is given to demonstrate how the framework serves for government policy making routines. Keywords: Macroeconomic policy making; decision support system; macroeconomic early warning. § Corresponding author. 335 Int. J. Info. Tech. Dec. Mak. 2009.08:335-359. Downloaded from www.worldscientific.com by KYUNGPOOK NATIONAL UNIVERSITY on 08/27/15. For personal use only. 336 X. Zhang et al.

No abstract

No abstract

It is necessary to build effectiveness infectious disease analysis methodology to avoid a large spreading of the disease. The factors playing role in epidemic come from different domain; moreover, their relationship is complex. Thus, it is very hard to mine the rule by single analysis. In this work, a total review is done to analyze the infected in the unit of years, which can provide a foundation to conclude the infectious rule. In order to finish this goal, Part Heuristic K-means based on Improved Grey Correlation Analysis is proposed. It uses improved grey correlation analysis to recognize the relevance among different diseases which has ability to guide the weight. Then, the year is partitioned into clusters based on distance function. It is found that the proportion of three degrees is respectively 21.4%, 28.6%, and 50%; the maximum of relevance is 0.888.

No abstract