The Application Of Decision Tree In Food Rotation Business

DSS has developed to many practical DSS systems involved operational research , decision-making and kinds of AI knowledge from initial system which could only manage the semi-structured problem via the assist manager in alternative technique , since it's appearance . The application of DSS involves many fields and has become one of the hotspots in information system domain . With the brought forward of the theory Data Warehouse and the informational found of Food Industry , it is mature to build up the Food Managing DSS of our country .Food is the base of the foundation . Reserved Food is the most important material foundation related to the safty of the economy of our country , the stabilization of the society , the pleasant living and working space for people , the sustainable development of our economy . At the same time , food has it's own life , which means that the reserved food must be whirled regular , otherwise the food will be musty and lose the action as reserved food .The paper begins with the practical application , with the purpose of solving the pratical decision problem in food management system , combines the conventional decision supported system with many other technique and method such as data warehouse , knowledge consequence , data mining , and implements the Decision System of Food Rotation Based on Data Warehouse .In this paper , several method of decision tree have been investigated , the basic knowledge of data mining and the development of classification method are introduced , summarizes the rules of food rotation using three decision tree algorithms : ID3 , C4.5 , CART , evaluates the summarized results using later stage pruning technique , shows the tree method of the rotation rules vividly with the graphical n-tree display technique . The reserved food in grain depot is judged by the summarized rules , the amount of goods of the grain depot which reaches the standard of rotation is early warned , and the decision directions of food rotation is accomplished . The unused rotation rules are stored as knowledge in the knowledge base , and are connected with later-stage Expert System to form the decision system of rotation integrated by Model Base , Method Base and Data Base .The Intelligent DSS of Grain (IDSSG) is a weighty science and technology item supported by the Department of Science and Technology of JiLin Province . The system employs Sybase ASE as the data warehouse server. The client point employs VC MFC 6.0 to implement decision Tree algorithms and PB programming to implement the rotation integrated circumstance. The system has three parts : rotation analyzing , data handling and maintenance , display of the result . The data maintenance module consists of three sub-functions : data creating , data operation , data maintenance . It is used in the data training of decision tree and the creation of test data . The rotation analysis module runs decision tree algorithms on training data , gain the needed rules , and improves the efficiency using pruning algorithm .This system is at the stage of prime research and trying out , and meet the need mainly . Still there are lots of work about making the system more perfect so as to meet the users' added need.