面向专业科技资源的服务定制推荐研究  

Research on Service Customization Recommendation for Professional Technology Resources

作　　者：邹宁 郭栋 Zou Ning;Guo Dong(Southwest Jiaotong University,Key Laboratory of Sichuan Province for Manufacturing Industry Chain Collaboration and Informatization Support Technology,Chengdu 610097;Beijing Machinery Industry Automation Research Institute Co.,Ltd.,Beijing 100120)

机构地区：[1]西南交通大学制造业产业链协同与信息化支撑技术四川省重点实验室,成都610097 [2]北京机械工业自动化研究所有限公司,北京100120

出　　处：《现代计算机》2021年第27期10-16,共7页Modern Computer

基　　金：国家重点研发计划“区域综合科技服务工程应用集成技术研究”(2017YFB1401400)。

摘　　要：由于专业科技资源服务平台提供的服务众多,且服务类别繁杂、用户难以搜寻到目标服务模块,这极大耗费了用户查找目标服务模块的时间,也降低了使用的满意度。针对此问题,本文使用服务评分模型处理用户行为数据,然后用改进修正余弦相似度的用户协同过滤算法,为目标用户推荐感兴趣的服务,进一步实现服务定制的功能模块。实验结果表明,改进后的协同过滤算法相比于传统算法,提高了服务推荐的质量和准确性。Due to the numerous services provided by the professional technology resource service platform and the complicated service categories,it is difficult for users to search for the target service module,which greatly consumes the time for users to search for the target service module and reduces user satisfaction.In response to this problem,this paper uses a service scoring model to process user behavior data,and then uses an improved user collaborative filtering algorithm with modified cosine similarity to recommend services of interest to target users,and further implement service customization functional modules.Experimental results show that the improved collaborative filtering algorithm improves the quality and accuracy of service recommendation compared with traditional algorithms.

关 键 词：专业科技资源 协同过滤 差异因子 最近邻居集 服务推荐 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]