Exchange Information And Data Mining Based On Association Rules

The data mining is a new kind of information processing technology。 It features in abstracting, transformation, analyzing the mass business data in the database and extracting the information helpful to making trade decisions. With data mining technology, we can find some unknown and valuable rules from the abundant trade data and help enterprises to apply these rules into the management consciously soas to improve the efficiency of it and enhance the capacity of competition.The association rules method is a main task of data mining. The rules express the relations between a group of objects in the database. A simple expression is:X=>Y.X is called the premise of the rules and Y is the result of the rules. A association rule is generally measured by two standers: support and confidence. Mining the association rules is to find the rules like this: its support and confidence are separately bigger than the given index of support and confidence. The most important step in the process of data mining is to find out all the frequent items whose support is no less than the given one. Apriori arithmetic is a kind of classical method used for searching for frequent items.The Futures is a kind of advanced form of exchange of commodities. With the development and ripe of market economy, its position and influence as a modern investment method are becoming more and more important in the current world economy. The futures companies did the settlement of the customer's business everyday and have accumulated a lot of business data. Thus, it is a meaningful task to take advantage of these data and mine the hidden rules behind it so as to help the companies do better in the aspect of developing more customers, guidance of business as well as risk control. Because of the factors such as unsoundness of china's future markets, the limit of future companies and the privacy of the data, there're nearly no research findings in this area in the public datum currently.This paper is focusing on the application of data mining technology in the future business and mainly does the research on the mining relation rules ofinformation in futures business. The main tasks are as follows:1, Do the research on Apriori arithmetic and improve its efficiency of it from two aspects. First, reduce the data required for scanning; Second, reduce the time for producing candicate items.2, Have an analysis of abundant trade data and select four indexes that can best reflect the trade situation. They are customer's fund, commission charge, risk and profit & loss; Classify these indexes according to their meaning and general customs in Futures companies in order to make it coincident to the practical demand of the management of Future companies.3> Design the system used for mining futures trade information based on association rules. The system adopts VB.NET as development tools and SQLSERVER as back database. It contains the functions such as abstracting and transformation of data, mining association rules and demonstration of results.4 > Mining from the trade information using the system and get some association rules between the four indexes. Have a special analysis on between the characteristics index-risk and other indexes. The results justify the experience of daily management as well as discover some mistakes in management that have a good guidance in future management for futures companies.From public materials and part of the research findings, there're no findings on relations between main indexes. This paper adopts Apriori arithmetic to do some research on this topic and gets correct and reasonable results that provide new ideas for future companies in taking full advantage of futures information and enlightenment of customer's management and control of risks.