《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (5): 1415-1423.DOI: 10.11772/j.issn.1001-9081.2024050743

• 2024年中国粒计算与知识发现学术会议 • 上一篇    

约简形式背景下的概念集构造及其推荐应用

陈昕1,2, 刘忠慧1,2, 闵帆1,2,3()   

  1. 1.西南石油大学 计算机与软件学院,成都 610500
    2.西南石油大学 机器学习实验室,成都 610500
    3.西南石油大学 人工智能研究院,成都 610500
  • 收稿日期:2024-06-04 修回日期:2024-07-13 接受日期:2024-07-23 发布日期:2024-08-26 出版日期:2025-05-10
  • 通讯作者: 闵帆
  • 作者简介:陈昕(2000—),男,四川德阳人,硕士研究生,主要研究方向:形式概念分析、机器学习
    刘忠慧(1980—),女,四川南充人,教授,硕士,CCF会员,主要研究方向:机器学习、形式概念分析、粗糙集
    闵帆(1973—),男,重庆人,教授,博士,CCF会员,主要研究方向:粒计算、多标签学习、主动学习。
  • 基金资助:
    国家自然科学基金资助项目(61772002);中央引导地方科技发展专项(2021ZYD0003);南充市校科技战略合作项目(23XNSYSX0062)

Concept set construction of reduced formal context and its recommendation application

Xin CHEN1,2, Zhonghui LIU1,2, Fan MIN1,2,3()   

  1. 1.School of Computer Science and Software Engineering,Southwest Petroleum University,Chengdu Sichuan 610500,China
    2.Lab of Machine Learning,Southwest Petroleum University,Chengdu Sichuan 610500,China
    3.Institute for Artificial Intelligence,Southwest Petroleum University,Chengdu Sichuan 610500,China
  • Received:2024-06-04 Revised:2024-07-13 Accepted:2024-07-23 Online:2024-08-26 Published:2025-05-10
  • Contact: Fan MIN
  • About author:CHEN Xin, born in 2000, M. S. candidate. His research interests include formal concept analysis, machine learning.
    LIU Zhonghui, born in 1980, M. S., professor. Her research interests include machine learning, formal concept analysis, rough set.
    MIN Fan, born in 1973, Ph. D., professor. His research interests include granular computing, multi-label learning, active learning.
  • Supported by:
    National Natural Science Foundation of China(61772002);Central Government Funds of Guiding Local Scientific and Technological Development(2021ZYD0003);Nanchong Municipal Government-Universities Scientific Cooperation Project(23XNSYSX0062)

摘要:

在形式概念分析(FCA)领域,概念集合的提出满足了真实环境的推荐需求;但目前概念集合生成方法缺乏有效的手段避免冗余属性的参与,这在一定程度上影响了概念生成的质量和效率,最终影响了推荐的效果。针对上述问题,提出形式背景属性约简算法(FCAR)、概念集构造算法(CSCA)以及基于概念集合的推荐算法(RACS)。首先,根据形式背景和评分矩阵设计属性兴趣度,并根据属性兴趣度阈值实现形式背景约简;其次,结合外延相似性与内涵兴趣度设计概念关键度作为启发信息,生成概念集合;最后,利用推荐置信度与推荐阈值得到概念集的推荐矩阵,从而针对目标用户实现个性化推荐。在11个数据集上对比了RACS与算法k最近邻(kNN)、基于项目的协同过滤(IBCF)、启发式概念集构造的组推荐(GRHC)、基于概念集的个性化推荐(CSPR)以及GreConD-kNN。实验结果表明,在6个常规数据集上,RACS在3个数据集上取得最高精确度和次高召回率,在4个数据集上取得最优F1值;特别是在3个较大规模的数据集上,与三种形式概念的推荐算法相比,RACS的推荐时间效率至少提升8倍。实验结果验证了RACS在推荐效果和推荐效率上的显著优势。

关键词: 形式概念分析, 约简形式背景, 属性兴趣度, 概念构造, 推荐系统

Abstract:

In the field of Formal Concept Analysis (FCA), the proposal of concept set satisfies the recommendation needs of real environments. However, current concept set generation methods lack effective means to avoid the inclusion of redundant attributes, which to some extent affects the quality and efficiency of concept generation, and ultimately the effectiveness of recommendations. To solve the above problem, a Formal Context Attribute Reduction algorithm (FCAR), a Concept Set Construction Algorithm (CSCA), and a Recommendation Algorithm based on Concept Set (RACS) were proposed. Firstly, the attribute interest degree was designed based on formal context and rating matrix, and formal context reduction was realized according to the threshold of attribute interest degree. Secondly, by combining extent similarity and intent interest degree, the concept criticality was designed as heuristic information to generate the concept set. Finally, the recommendation matrix of the concept set was obtained using the recommendation confidence and recommendation threshold, enabling personalized recommendation for the target user. RACS was compared with algorithms including k-Nearest Neighbor (kNN), Item-Based Collaborative Filtering (IBCF), Group Recommendation based on Heuristic Concept set (GRHC), Concept Set based-Personalized Recommendation (CSPR), and GreConD-kNN on 11 datasets. In experiments on six standard datasets, RACS achieves the highest accuracy and the second highest recall on three datasets, and achieves the best F1 score on four datasets. Especially on three larger-scale datasets, compared to formal concept recommendation algorithms, RACS has recommendation time efficiency improved by at least eight times. Experimental results validate the significant advantages of RACS in recommendation effects and efficiency.

Key words: Formal Concept Analysis (FCA), reduced formal context, attribute interest degree, concept construction, recommendation system

中图分类号: