Research On Decision Support Technique Based On Data Warehouse

Since the decision support techniques came into the world, a lotof challenge researches are brought out. A large number of researchfruit such as machine study, statistics, database system are used in thisfield directly. A single system cannot solve all the problem, we mustintegrate all these system to make decision support. From a certainmeaning, the decision support techniques have been become a newand specialty research field, and it have been used in a lot of otherknowledge area, the decision support techniques have promoted allthese knowledge, it has produced more and more important influencein the storage, management and discover of the knowledge.This paper general introduces the development of the decisionsupport techniques, the techniques used in the system and the researchsteps. The decision support techniques can analyze the dataautomatically, sum up reasoning and associating these data, findingthe inner relationship between these data, digging up the t potentialpattern which have important effect with information forecast anddecision action. The decision support techniques involved in database,artificial intelligence, machine study, nerve calculation and statisticalanalyze, it has brought up a new phase in the management andapplication of data.This paper researched the design flow of the data warehouse anddata mining model, the specialty relevant key techniques such asanalyze decision tree, faint logic, evolve calculation, nerve net,regulation sum up and statistical method, the important problem andhandle point about the model carry out and quality analyze, theresearch fruit has been carried out in the postal CRM system. Thispaper also analysed and managed the quality problem which an bepossibly occurred in the data miningging process. After the practice,the designed data warehouse and data mining model have acquiredpreferably effect in the postal business data mining.