社会化标注系统中个性化信息推荐多维度融合与优化模型研究

详细信息    查看全文 | 推荐本文 |

英文篇名：Research on the Multi-dimensional Fusion and Optimization Model of Personalized Information Recommendation in Social Tagging System 作者：武慧娟 ; 孙鸿飞 ; 金永昌 英文作者：Wu Huijuan;Sun Hongfei;Jin Yongchang;School of Economics and Management,Northeast Dianli University;School of Management,Jilin University; 关键词：个性化信息推荐 ; 社会化标注 ; 多维度融合 ; 优化模型 英文关键词：personalized information recommendation;;social tagging system;;multi-dimensional fusion;;optimization model 中文刊名：XDQB 英文刊名：Journal of Modern Information 机构：东北电力大学经济管理学院;吉林大学管理学院; 出版日期：2018-12-27 出版单位：现代情报 年：2019 期：v.39;No.331 基金：教育部人文社会科学规划项目“社会化标注系统中个性化信息推荐多维度融合与优化模型研究”(项目编号:15YJC870024);; 吉林省教育厅“十三五”社会科学研究规划项目“基于用户认知的多源融合个性化微阅读用户画像模型研究”(项目编号:JJKH20190715SK) 语种：中文; 页：XDQB201901006 页数：7 CN：01 ISSN：22-1182/G3 分类号：38-43+86

摘要

[目的/意义]在社会化标注系统自组织运行的基础上,构建个性化信息推荐的多维度融合与优化模型,进而在大数据环境下,为用户提供精准的个性化信息推荐服务,从而进一步丰富个性化信息推荐的理论体系以及拓展个性化信息推荐的研究方法。[方法/过程]首先,对每一种个性化信息推荐方法的优点和不足进行深入分析;然后,将基于图论(社会网络关系)、基于协同过滤以及基于内容(主题) 3种个性化信息推荐方法进行多维度深度融合,构建个性化信息推荐多维度融合模型;最后,对社会化标注系统中个性化信息推荐多维度融合模型进行优化,从而解决个性化推荐过程中用户"冷启动"、数据稀疏性和用户偏好漂移等问题。[结果/结论]通过综合考虑现有的基于图论(社会网络关系)、基于协同过滤以及基于内容(主题)的个性化信息推荐方法各自的贡献和不足,实现3种方法之间的多维度深度融合,并结合心理认知、用户情境以及时间、空间等优化因素,最终构建出社会化标注系统中个性化信息推荐多维度融合与优化模型。
        [Purpose/Significance] On the basis of the self-organizing operation of the social tagging system,the multi-dimensional fusion and optimization model of personalized information recommendation was constructed to provide accurate personalized information recommendation service for users under the big data environment,thus further enriching the theoretical system of personalized information recommendation and expanding the research methods of personalized information recommendation. [Method/Process] Firstly,the advantages and disadvantages of each personalized information recommendation method were analyzed in depth. Then, three personalized information recommendation methods based on graph theory( social network relationship),collaborative filtering and content-based( topic-based) were fused in depth to construct a personalized information recommendation multi-dimensional fusion model. Finally,the multi-dimensional fusion model of personalized information recommendation in social annotation system was optimized to solve the problems of cold start,data sparsity and users' preference drift in personalized recommendation process. [Result/Conclusion] By considering the respective contributions and shortcomings of graph theory( social network relationship),collaborative filtering and content-based( topic-based) personalized information recommendation methods,the multi-dimensional and deep fusion among the three methods was realized. Finally,a multi-dimensional fusion and optimization model of personalized information recommendation in social annotation system was constructed by combining psychological cognition,user context,time and space.

引文

[1] Resnick P,Varian H R. Recommendation Systems[M]. ACM,1997.
    [2] Balabanovic M,Shoham Y. Fab:Content-based,Collaborative Recommendation[J]. Communications of the ACM, 1997, 40(3):66-72.
    [3]张琳,闫强.基于管理和消费者行为视角的个性化推荐研究与展望[J].北京邮电大学学报:社会科学版,2016,18(6):24-30.
    [4] Garcin F,Zhou Kai,Faltings B,et al. Personalized News Recommendation Based on Collaborative Filtering[C]//Proceedings of the 2012 IEEE/WIC/ACM International Joint Conferences on Web Intelligence and Intelligent Agent Technology. Washington,DC,USA:IEEE,2012:437-441.
    [5] Darvishy A,Ibrahim H,Mustapha A,et al. New Attributes for Neighborhood-based Collaborative Filtering in News Recommendation[J]. Journal of Emerging Technologies in Web Intelligence,2015,7(1):13-19.
    [6]张亮,赵娜.改进的协同过滤推荐算法[J].计算机系统应用,2016,25(7):147-150.
    [7] Hu J,et al. Personalized Tag Recommendation Using Social Influence[J]. Journal of Computer Science and Technology,2012,27(3):527-540.
    [8]唐晓波,张昭.基于混合图的在线社交网络个性化推荐系统研究[J].情报理论与实践,2013,36(2):91-95.
    [9] Mandl M,Felfering A,Teppan E,et al. Consumer Decision Making in Knowledge-based Recommendation[J]. Journal of Intelligent Information Systems,2011,37(1):1-22.
    [10]姜信景,齐小刚,刘立芳.个性化信息推荐方法研究[J]智能系统学报,2018,13(2):189-195.
    [11]刘平峰,聂规划,陈冬林.基于知识的电子商务智能推荐系统平台设计[J].计算机工程与应用,2007,43(19):199-201.
    [12]吴兵,叶春明.基于效用的个性化推荐方法[J].计算机工程,2012,38(4):49-51.
    [13] Carrer-Neto W,Hernández-Alcaraz M L,Valencia-García R,et al. Social Knowledge-based Recommender System. Application to the Movies Domain[J]. Expert Systems with Applications,2012,39(12):10990-11000.
    [14]覃容,陈建峡.基于协同过滤和内容的用户需求混合推荐算法[J].企业技术开发,2018,37(2):67-69.
    [15]武慧娟,徐宝祥.社会化标注系统自组织演化机理研究[J].图书情报工作,2014,58(2):132-137.
    [16]武慧娟,崔金栋,徐宝祥.基于多层级耗散的社会化标注系统序化研究[J].情报科学,2015,33(8):25-29.