基于大数据的人工智能跨境电商导购平台信息个性化推荐算法

传统算法计算与存储开销大,影响推荐结果准确性,不适于含大规模数据的人工智能跨境电商导购平台信息的个性化推荐的。为此,通过大数据技术研究人工智能跨境电商导购平台信息个性化算法,使得大数据技术在Hadoop平台实现,通过Map将任务分解成多个任务,采用Reduce将分解后多任务处理结果集合在一起,获取最终处理结果。通过两个MapReduce与一个map对平台中用户偏好获取算法进行并行化处理。针对用户偏好,通过关联规则挖掘获取和用户偏好相符的商品,推荐给用户。结果表明:所提算法推荐准确率、召回率和平均精度均高于其他算法;所提算法推荐商品符合用户偏好;所提算法推荐商品信息点击率与转换率最优。可见所提算法推荐精度高,推荐商品信息可满足用户偏好,应用性强。

Personalized recommendation algorithm based on big data for artificial intelligence cross border e-commerce shopping platform

In order to solve the problem that traditional algorithms spend too much on computation and storage, which affects the accuracy of recommendation results, it is not suitable for the personalized recommendation of information of AI cross-border e-commerce shopping platform with large-scale data. Through big data technology, the information personalization algorithm of AI cross-border e-commerce shopping platform is studied. Big data technology is implemented on Hadoop platform. Tasks are decomposed into multiple tasks by Map, and the decomposed multi-task processing results are assembled by Reduce to obtain the final processing results. We parallelize the user preference acquisition algorithm in the platform through two MapReduce and one map. According to user preferences, the products matching user preferences are obtained by mining association rules and recommended to users. The results show that the recommendation accuracy, recall rate and average accuracy of the proposed algorithm are higher than those of other algorithms; the proposed algorithm recommends goods in line with user preferences; the proposed algorithm recommends goods with the best click-through rate and conversion rate. We can see that the proposed algorithm has high recommendation accuracy, and the recommended commodity information can satisfy user preferences and has strong applicability.