An Apriori Improved Algorithm Of Data Mining Based On Graph And Implementation Of A Data Mining System

Recently , with the rapid development of the information industry , the data stored in the information system has become more and more larger . The necessary of data management has become mining the interested regular from the history data. Then the regular is used to guide the future behavior. Because of this, data mining technology came into being.While, faced the large data, the efficiency of the traditional data mining algorithm is needed to be increased. So we need to do some analysis of association rule generation algorithm of data mining technology and proposed the improved algorithm based on the graph mining. On the light of the representation of the vertical data format, the storage of database in apriori algorithm has been improved.Based on graph mining, the mining of the candidate frequent item set has been improved greatly, which achieved mining all candidate frequent item set one time. This improved algorithm increased data mining efficiency apparently in mining longer item set and larger transaction data. So, it is appropriate to mining longer transaction and larger data set using of this algorithm.The theory is needed to verify by practice. So, a data mining system is designed based on the improved apriori algorithm. We use this system to check out the more excellent performance of the improved apriori algorithm comparing with traditional apriori algorithm. Using of this system, it is proved that improved apriori algorithm is more excellent than the traditional apriori algorithm.