摘要
本文针对基于内容的论文推荐系统推荐论文在质量与时效性等方面存在的不足,提出了一种基于多维计量权重的学术论文推荐模型。结合内容个性化推荐的技术方法和文献计量学,从论文的期刊影响力、作者贡献度、文献老化指数等多个维度对论文推荐过程进行优化,并采用层次分析法确定各维度指标的权重,最终生成个性化的论文推荐列表。模型实证效果较好,可为优化信息服务方式、减轻研究人员文献筛选负担提供新思路。
The paper introduces an academic paper recommendation model based on multidimensional metric weights to address the deficiencies in quality and timeliness of content-based paper recommendation systems.By in-tegrating the techniques of personalized content recommendation and bibliometrics,the paper recommendation process is optimized across multiple dimensions including journal influence,author contribution,and literature aging index.The Analytic Hierarchy Process is employed to ascertain the weights of indicators within each dimension,culminating in the production of a personalized paper recommendation list.The model demonstrates effective empirical perform-ance,offering fresh insights for enhancing information service strategies and alleviating the literature screening bur-dens on researchers.
作者
迟玉琢
李金灿
Chi Yuzhuo;Li Jincan
出处
《山东图书馆学刊》
2024年第5期39-48,共10页
The Library Journal of Shandong
关键词
学术论文推荐
内容推荐系统
文献计量学
多维指标
Academic paper recommendation
Personalized content recommendation
Bibliometrics
Multi-dimensional indicators
