• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Wireless Communications and Mobile Computing Conference >Application of Apriori Association Rules Algorithm to Data Mining Technology to Mining E-commerce Potential Customers
【24h】

Application of Apriori Association Rules Algorithm to Data Mining Technology to Mining E-commerce Potential Customers

机译：Apriori关联规则算法在挖掘电子商务潜在客户的数据挖掘技术中的应用

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Because of the massive amount of source data, certain data relationships are often hidden in these data, and there are certain trends in the data. For e-commerce, the prediction and estimation of these data trends are particularly important. Providing reasonable forecasts will help companies make effective decisions about the current situation and provide customers with appropriate services, thereby helping companies seize customers to generate more commercial profits. This paper analyzes the shortcomings of the Apriori algorithm and proposes an association rule mining technology to optimize the shortcomings of the Apriori algorithm. In the experiment, the test code was written in the Web environment, and the optimized algorithm and the Apriori algorithm were used for data mining on the same number of itemsets under the same support conditions. Experimental data shows that the improved algorithm has a certain improvement compared to the Apriori algorithm, and the performance is more obvious when the support is low. The experimental results show that when the support is 0.1, 0.2, 0.3, 0.4, 0.5, the improved Apriori algorithm is 12s, 17s, 12s, 8s, and 6s more than the traditional Apriori algorithm. In the case of a high degree of support, the improved algorithm has no obvious advantages. This is because the number of times the algorithm is executed is low, and the auxiliary table participates in the mining algorithm less frequently.
机译：由于源数据量大量，某些数据关系通常隐藏在这些数据中，并且数据存在某些趋势。对于电子商务，这些数据趋势的预测和估计尤为重要。提供合理的预测将有助于公司对目前情况做出有效的决策，并为客户提供适当的服务，从而帮助公司抓住客户以产生更多的商业利润。本文分析了APRiori算法的缺点，提出了一个关联规则挖掘技术，以优化APRiori算法的缺点。在实验中，测试代码被写入Web环境，并且优化算法和APRiori算法用于在相同的支持条件下对相同数量的项目集进行数据挖掘。实验数据表明，与APRiori算法相比，改进的算法具有一定的改进，并且当支撑率低时，性能更加明显。实验结果表明，当支撑件为0.1,0.2,0.3,0.4,0.5时，改进的APRiori算法是12S，17S，12S，8S和6S，比传统的APRIORI算法多。在高度支持的情况下，改进的算法没有明显的优势。这是因为算法被执行的次数低，并且辅助表频繁地参与挖掘算法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号