关键词: 数据挖掘, 隐私保护, 增量更新

Abstract: Abstract: With the arrival of large data age, efficiency in mining had become the bottleneck of frequent pattern mining algorithms for privacy-preserving, and a small amount of database updates also need to re-run the original algorithm, often resulting in a great waste of mining results before, an update model must be added to the mining algorithm to improve its efficiency. In order to solve the problem that the GRRPH algorithm needs multiple database scans during the running time and counting several times, this paper proposed the improved algorithm BRRPH of GRRPH. BRRPH algorithm uses bitmap technology to represent transaction in the database, the results of the experiment show that the improved algorithm has higher efficiency than the RRPH algorithm. Aimed at the frequent or occasional additions to the database will make some of the existing strong association rules invalid or weak rules become strong rules, this paper introduced an incremental update model, and then proposed the IBRRPH algorithm. The results of the experiment show that the improved algorithm has higher efficiency than the RRPH algorithm.

Key words: data mining, privacy-preserving, incremental update