Study In Incremental Mining Association Rules And Application On FOQA

Data mining is a process to revealing latent and interesting knowledge from massive data, and an effective approach to solve the problem of"rich data and poor Knowledge". Association rules mining can reveal interesting correlations between item sets from massive data. It is an important subject of data mining and is widely used in real life.This paper mainly focuses on the construction of incremental mining association rules system and application on FOQA. Traditional methods of data mining face"massive data", is only"in the new parameters under re-implementation of a constraint", and there is no any relationship with last data mining process and results, causing a waste of resources. How to carry out data mining on the data sets updated is a problem which researchers can not avoid."Incremental"mining algorithm provides a feasible way to solve mining variable data sets.The main work of this paper includes implementing classic mining association rules algorithm, and comparing them in experiment; with the attribute reduction based on rough set, improving the RSQR algorithm and make it can do attribute reduction incrementally; improving the H-mine algorithm, purPOSe the incremental mining item sets based on H-mine and XML and applying the process of data pretreatment and data mining to FOQA. Through the study of this paper, with the help of attribute reduction the incremental mining item sets based on H-mine and XML shows a high performance in the field of Chinese text classification and is supPOSed to have a good application prospect.