摘要
为了加强保险业务员业务能力,提高被推荐保险产品的成交率,基于经典关联规则挖掘算法Apriori,提出了一种改进算法Apriori based on difference set(DS Apriori)来计算符合条件的保险产品。贡献包括:(1)通过以键值方式重新组织数据、根据时序划分数据集、降低迭代搜索的次数等方式来提高算法执行效率;(2)通过以客户为主键聚合事务数据来挖掘潜在关联规则,实现精准推送。实验表明所提出的DS Apriori算法执行效率优于Apriori等算法,可以给保险业务员提供业务指导,精准推荐最符合推荐条件的保险产品。
In order to enhance the business capability of insurance salesmen and improve the turnover rate of recommended insurance products,an improved algorithm Apriori based on difference set(DS Apriori)is proposed to calculate the eligible insurance products based on the classical association rule mining algorithm Apriori.The contributions include(1)improving the algorithm execution efficiency by reorganizing the data in key-value way,dividing the data set according to the time sequence,and reducing the number of algorithm iterations;(2)mining the potential association rules by aggregating the transaction data with the customer as the main key to achieve accurate pushing.The experiments show that the proposed DS Apriori algorithm performs better than algorithms such as Apriori.It can provide business guidance to insurance salesmen and accurately recommend the insurance products that best meet the recommended conditions.
作者
朱天宇
谭文安
ZHU Tianyu;TAN Wen’an(School of Computer and Information Engineering,Shanghai Polytechnic University,Shanghai 201209,China)
出处
《上海第二工业大学学报》
2022年第2期172-176,共5页
Journal of Shanghai Polytechnic University
基金
国家自然科学基金项目(61672022,U1904186)
上海第二工业大学电子信息类专业硕士协同创新平台建设项目资助。
关键词
关联规则
差集
APRIORI算法
保险推荐
association rule
difference set
Apriori algorithm
insurance recommendation