基于MapReduce的频繁项目集挖掘算法在煤炭销售系统中的研究

煤炭系统中,往往希望分析不同煤炭产品购买之间的关联规则,并通过一定的关联性推荐煤炭商品,有助于购买者购买并取得更高的销售量。发掘频繁项目集是关联规则中经常用到的关键技术。随着煤炭系统数据库中信息的增多,原有的频繁项目集挖掘算法无法快速高效地完成频繁项目的挖掘。针对海量数据信息频繁项目集挖掘问题,提出了分布式频繁项目集挖掘算法,该算法是基于MapReduce分布式计算框架,能够高效地完成数据库中的频繁项目挖掘工作。通过实验结果证明该算法具有很高的效率及可扩展性。

Research on Frequent Item-set Algorithm in Coal System Based on MapReduce

In coal system, we usually hope to analyze the association rules, and recommend the products according to the association rule, which is helpful for the customers to buy and get higher sales volume. Discovering frequent item-sets is a key technology which is usually used in association rules. With the increase of information in database of coal system, traditional frequent item-set algorithm could not complete the mining work of frequent item-sets. Focusing on large scale frequent item-sets mining problem, we propose distributed frequent item-sets mining algorithm, and this algorithm is based on MapReduce distributed computing framework, and it can complete frequent itemset mining work highly effectively. The experimental results prove the algorithm proposed has high efficiency and scalability.