摘要
针对在数字图书馆中刻画教育者和学习者图书偏好的文本和网络数据不精确、不一致和不通用问题,提出一种基于模糊本体和遗传算法的推荐系统框架。将模糊逻辑引入领域本体,处理在图书领域中的模糊信息,采用遗传算法对图书的特征进行权重优化,通过聚类算法缩小推荐搜索空间,达到精细化的推荐结果。实验结果表明,该方法有效解决了冷启动、数据稀疏性问题,解决了本体推荐图书的不确定性和主观判断问题,与传统方法相比精度和泛化能力有所提高。
Aiming at the inaccuracy, inconsistency, and non-generalization of text and network data that portray educators and learners’ book preferences in digital libraries, a recommendation system framework based on fuzzy ontology and genetic algorithms was proposed. Fuzzy logic was introduced into the domain ontology to process the fuzzy information in the field of books, and the genetic algorithm was used to optimize the weight of the book ’s features. The recommended search space was reduced using the clustering algorithm to achieve refined recommendation results. Experimental results show that the proposed method solves not only the problems of cold start and data sparseness, but the uncertainty and subjective judgment problems of ontology recommendation books. Compared with the traditional methods, the accuracy and generalization ability are improved.
作者
郭斯檀
潘广贞
赵利辉
郭雁蓉
GUO Si-tan;PAN Guang-zhen;ZHAO Li-hui;GUO Yan-rong(Software School,North University of China,Taiyuan 030051,China)
出处
《计算机工程与设计》
北大核心
2019年第3期834-838,共5页
Computer Engineering and Design
基金
山西省应用基础研究计划--山西省面上自然基金项目(201701D121066)
山西省高等学校科技创新基金项目(2017156)
关键词
模糊本体
遗传算法
推荐
聚类
数字图书馆
fuzzy ontology
genetic algorithm
recommendation
clustering
digital library