基于AIS算法和Lambda的医学图书馆推荐系统构建  

Construction of medical library recommendation system based on AIS algorithm and lambda

作　　者：张炯娜 王若瑶 何晓茜 ZHANG Jiong-na;WANG Ruo-yao;HE Xiao-xi(Shanghai Chest Hospital(School of Medicine,Shanghai Jiaotong University)Ministry of Science and Education,Shanghai 200030,China)

机构地区：[1]上海市胸科医院(上海交通大学医学院附属胸科医院)科教部,上海200030

出　　处：《信息技术》2024年第12期80-85,92,共7页Information Technology

基　　金：中国图书馆学会医学图书馆分会科研资助项目(YX2021Y09)。

摘　　要：文中提出了一种基于人工免疫系统(AIS)算法和Lambda架构的医学图书馆智能推荐系统。该系统通过提取用户的浏览历史、阅读习惯等特征,构建用户兴趣模型。然后基于AIS算法的免疫克隆选择机制产生个性化的推荐结果。同时,系统采用Lambda架构实现了实时推荐功能。实验结果显示,与基于协同过滤的推荐系统相比,该系统的推荐精度可达73%,召回率达72%。实时推荐的延迟在毫秒级,满足图书馆用户的实时性需求。该系统为用户提供了及时、个性化的医学图书馆推荐服务,提高了图书馆利用率和读者满意度。This paper proposes an intelligent recommendation system for medical libraries based on Artificial Immune System(AIS)algorithm and Lambda architecture.The system constructs a user interest model by extracting the user’s browsing history,reading habits and other features.Then it generates personalized recommendation results based on the immune clone selection mechanism of the AIS algorithm.Meanwhile,the system adopts Lambda architecture to realize the real-time recommendation function.The experiment results show that compared with the recommendation system based on collaborative filtering,the recommendation precision of this system can reach 73%,and the recall rate reaches 72%.The delay of real-time recommendation is in milliseconds,which meets the real-time demand of library users.This system provides users with timely and personalized medical library recommendation services,and improves library utilization and reader satisfaction.

关 键 词：AIS算法 Lambda架构 医学图书馆智能推荐系统 模型训练 个性化推荐 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]