针对目前传统的Apriori算法对硬件要求较高且运算效率低下的情形，提出将经典的数据挖掘关联规则算法Apriori移植到云计算平台，并结合MapReduce机制进行海量数据挖掘，有效地解决了传统Apriori算法存在的瓶颈问题以及对硬件要求高的依赖。通过数据和节点对比实验共同验证了移植后的Apriori算法的运算效率比传统的Apriori算法提高了许多倍，且随着数据量和节点数的增加效果愈发明显。由于改良后的Apriori算法具有高效性和可行性，这将为解决当前大数据挖掘问题提供了一种全新的、有效的解决方案，并且这一结论还可为其他数据挖掘算法的移植提供可靠的参考。%For the existing problems of the traditional Apriori algorithm on higher hardware requirements and inefficiency, the classical data mining association rules algorithm Apriori transplanted into cloud computing platform is proposed, and combined with the MapReduce mechanism for massive data mining, it can effectively solve the bottleneck problems of conventional Apriori algorithm and the dependence on high-performance computer. By comparing the experimental data and nodes verify Apriori algorithm transplantation efficiency many times than the traditional Apriori algorithm, and with the amount of data and the number of nodes increase the effect more obvious. Because the Apriori algorithm after improvement is effective and feasible, which will solve the current problem of large data mining provides a new, effective solutions, and this conclusion can provide reliable reference for other data mining algorithm transplantation.
展开▼
