A Study On Algorithms Of Association Rules In Data Mining

The society is developing swiftly. The worldwide application of the computer and information technology has been changing people's life, and the enterprise's operation method, too. In an environment of severity market competition, it is necessary for the enterprise to apply the advanced computer and information technology to win the market. Therefore, the DSS is wildly applied,the data Ming and data market technology, which is the basement of it, are grown up widely too.Data Ming is a most active and new research in the field of Artificial Intelligence and field of Database in the world. It's goal is to satisfy the users' need, process the great amount of original data and identify the important and interesting patterns in It. And at last expressing them In the form of knowledge which people can recognize at their first sight. Data Ming is a kind of process that reveals potential useful knowledge from massive data. It Is an efficient way to tackle"Data Rich and information Poor". Association rules are an important aspect of research of Data Ming.The Apriori algorithms is a kind of good algorithms to mine association Rules. But because it has to scan the big database many times, it will take a long time to run it. This dissertation has shown a new algorithm after building the data market. The new algorithms produces a more small amount of candidate patterns when It looks for the frequent patterns of the DB, using theory that the Parent pattern of the non_frequent pattern is non_freqent pattern. So it can reduce the scale of the candidate_frequent patterns greatly. It can be easily implemented, and can improve the efficiency greatly by deleting those useless records.The algorithms has shown advantages such as: 1)High efficiency. Comparing with the Apriori algorithms,it can reduce the times to scan the DB. 2)Easy to realize. It is easy to realize the algorithms by temporary tables in PowerBuilder.