Study On Algorithm Of Mining Association Rules And Its Application In CRM

Data Mining or Knowledge Discovery emerged in the late 1980s has become a hotspot in the fields of artificial intelligence and database technology. Data mining has its wide application prospect and is expected to continue to flourish in the new millennium. R. Agrawal etc. first put forward the issue of mining association rules in 1993. Now it has been an significant content of data mining and so draws attention of many researchers.After we have studied existing algorithms of mining association rules such as Apriori, Incremental Updating algorithm etc. , two problems are found. One of the problems is that most of the algorithms must scan the whole large database when new data are added to it. So it will make the discovering frequent item sets very slow. Another problem is that the item sets which includes new items will be often regarded as unfrequent item sets even if they happened frequently in new data set because the support of the item sets is calculated based on the whole database. So the association rules come from above frequent item sets can't reflect the recent business activity.Having known the problems of existing algorithms, first I bring forward a new concept-sensitivity to measure how much the algorithms thinks of the new items which appeared in the new data set. Then on the one hand a parameter c(1≤c≤ ∞) is introduced for improving the efficiency of the algorithms. Discovering frequent item sets are based on the frequent item sets(support≥minsup/c) of the old data set and the new data set, not on the whole database. On the other hand thinking of the sensitivity, if frequent item sets only appeared frequently in the new data set, they will be directly added to the frequent item sets. In addition improved algorithm is compared with Incremental Updating Algorithm by experiment. At last the application of association rules in CRM based on data warehouse is discussed.