基于MapReduce的民航收益漏洞规则提取研究

Research of Civil Aviation Revenue Leakage Rules Extraction Based on MapReduce

在民航业务中,旅客订座后却不能如期登机(NOSHOW),最终会导致航班上座位的虚耗。针对海量的CKI(CheckIn,离港)数据,提出了一种分布式环境下基于NOSHOW约束的收益漏洞规则提取方案。首先对基于项约束条件的Apriori算法进行了论证,提出分布式环境下基于项约束的频繁项集的计算模型。实验结果表明,该算法能避免大量无用频繁项集的生成,加快Apriori算法迭代的速度。实验取得了不同置信度下的关联规则,可以为航空公司制定不同旅客的出票时限,提供智能化的决策支持。

In the aviation business, passengers cannot be boarding （NOSHOW） after booked, which will waste the flight seat waste. Faced with the humongous Check - In data, a rules extraction computational model is proposed based on NOSHOW constraints in distributed environment. Experiment result shows that this algorithm can avoid the generation of a large number of useless frequent item - sets, and speed up the iteration of the Apriori algorithm. Experiment result obtains association rules under different confidence, which can provide intelligent decision support for airlines to formulate different ticketed time.